finetuning_llama2_7b_1gpu.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "595a4b9b3e11459b939041adf43e4f5b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_d05c3fd567d24d1e82bfc81a5d390761",
              "IPY_MODEL_42c0706b4b70423ba54bac9ef18e49e2",
              "IPY_MODEL_2c1edac5c0dd4acb96bf24c0ae53d157"
            ],
            "layout": "IPY_MODEL_133263df5e8348c6ab1146c67cd8750d"
          }
        },
        "d05c3fd567d24d1e82bfc81a5d390761": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_adee6ed3b99e47ee8891748798d61dfd",
            "placeholder": "​",
            "style": "IPY_MODEL_b16ba452c3a94fd6a9eebfa605637c44",
            "value": "Downloading (…)fetensors.index.json: 100%"
          }
        },
        "42c0706b4b70423ba54bac9ef18e49e2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_38135f188f2944ef97a3d2fc2d7a0f65",
            "max": 26788,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_c4bf8f58d64544b5a237b26c26d2853d",
            "value": 26788
          }
        },
        "2c1edac5c0dd4acb96bf24c0ae53d157": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_52831548f39744aa8ef6388d6bafead5",
            "placeholder": "​",
            "style": "IPY_MODEL_1fbea419c6b640698de5625d1afa89ac",
            "value": " 26.8k/26.8k [00:00&lt;00:00, 1.25MB/s]"
          }
        },
        "133263df5e8348c6ab1146c67cd8750d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "adee6ed3b99e47ee8891748798d61dfd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b16ba452c3a94fd6a9eebfa605637c44": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "38135f188f2944ef97a3d2fc2d7a0f65": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c4bf8f58d64544b5a237b26c26d2853d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "52831548f39744aa8ef6388d6bafead5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1fbea419c6b640698de5625d1afa89ac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "51f7cb00a1bd437386d09109d390c3c0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_576dcb8e5d514455a577d24380137869",
              "IPY_MODEL_68b38196f0ee4562b281f692ff733701",
              "IPY_MODEL_748a76f4ddba45bc99c2d1510875a892"
            ],
            "layout": "IPY_MODEL_4e8f9325fed94f9898fe20348cd9f90a"
          }
        },
        "576dcb8e5d514455a577d24380137869": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c636953cbc9e4f5abc7df0e2ae360d64",
            "placeholder": "​",
            "style": "IPY_MODEL_fe5c12a50dd6472d848225f4ddaf1093",
            "value": "Downloading shards: 100%"
          }
        },
        "68b38196f0ee4562b281f692ff733701": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ddd182af26ac418fbb1f473c9cf61b27",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_9afe6667694d4a23abe0d0754f8d3662",
            "value": 2
          }
        },
        "748a76f4ddba45bc99c2d1510875a892": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e1c2c41b88ef4198827adb3511dde821",
            "placeholder": "​",
            "style": "IPY_MODEL_d7ea9730a93e47c3a16d2c775b086495",
            "value": " 2/2 [01:22&lt;00:00, 37.58s/it]"
          }
        },
        "4e8f9325fed94f9898fe20348cd9f90a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c636953cbc9e4f5abc7df0e2ae360d64": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fe5c12a50dd6472d848225f4ddaf1093": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ddd182af26ac418fbb1f473c9cf61b27": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9afe6667694d4a23abe0d0754f8d3662": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "e1c2c41b88ef4198827adb3511dde821": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d7ea9730a93e47c3a16d2c775b086495": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6c4dbf8ac9cf48acb7c544ace6216901": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_9ba09e0c332149babfd471b32195f260",
              "IPY_MODEL_ab3e60aa968f467096aaa7a6541f329a",
              "IPY_MODEL_4bb1758031a84760a829959fbe66fd66"
            ],
            "layout": "IPY_MODEL_5c61d3aaf6b34808bdcefd511b460448"
          }
        },
        "9ba09e0c332149babfd471b32195f260": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f994af8718cd4afcbac161cc7a2bc2c2",
            "placeholder": "​",
            "style": "IPY_MODEL_f81ab7826c2f4312943b2b04ae3652a1",
            "value": "Downloading (…)of-00002.safetensors: 100%"
          }
        },
        "ab3e60aa968f467096aaa7a6541f329a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e794e79ed35c4e82a81d6366c9b15dbf",
            "max": 9976578928,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_146ba63c78d84a83b5dfc008ecc3f7cc",
            "value": 9976578928
          }
        },
        "4bb1758031a84760a829959fbe66fd66": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1785d0af5fe842b890549c697a9d96e3",
            "placeholder": "​",
            "style": "IPY_MODEL_622ca9d2d57d4fc7a7f4ab0cb2c9d16d",
            "value": " 9.98G/9.98G [01:02&lt;00:00, 260MB/s]"
          }
        },
        "5c61d3aaf6b34808bdcefd511b460448": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f994af8718cd4afcbac161cc7a2bc2c2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f81ab7826c2f4312943b2b04ae3652a1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e794e79ed35c4e82a81d6366c9b15dbf": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "146ba63c78d84a83b5dfc008ecc3f7cc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "1785d0af5fe842b890549c697a9d96e3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "622ca9d2d57d4fc7a7f4ab0cb2c9d16d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "7923a48bac9c49479cdab7f1a9d3caa3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_3d4d6cd6c8854e20b582e48cb2952f07",
              "IPY_MODEL_fcb2c53b52d9459289f7cbb270fcd0a2",
              "IPY_MODEL_5fd064b4e8254165bbbd3c32ecdead75"
            ],
            "layout": "IPY_MODEL_a1a54ff6d5284a5e9da499b2b0289a28"
          }
        },
        "3d4d6cd6c8854e20b582e48cb2952f07": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_29a3396d6926445eae50587a8d813794",
            "placeholder": "​",
            "style": "IPY_MODEL_f64c980723dd479c88058293efe781d0",
            "value": "Downloading (…)of-00002.safetensors: 100%"
          }
        },
        "fcb2c53b52d9459289f7cbb270fcd0a2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e8ddcf1677ac43d8868933ace87eaaa0",
            "max": 3500297344,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_cf92a6322d37402d9afcd46eda10f6e4",
            "value": 3500297344
          }
        },
        "5fd064b4e8254165bbbd3c32ecdead75": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3cb0685e40bb4de29ff7fb6e2d9b9d16",
            "placeholder": "​",
            "style": "IPY_MODEL_ba470fbdc9b44ddfaf52f51ffd11c651",
            "value": " 3.50G/3.50G [00:19&lt;00:00, 213MB/s]"
          }
        },
        "a1a54ff6d5284a5e9da499b2b0289a28": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "29a3396d6926445eae50587a8d813794": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f64c980723dd479c88058293efe781d0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e8ddcf1677ac43d8868933ace87eaaa0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cf92a6322d37402d9afcd46eda10f6e4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "3cb0685e40bb4de29ff7fb6e2d9b9d16": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ba470fbdc9b44ddfaf52f51ffd11c651": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "91b7da766f844564ad2c54c6da09def9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_a3e0192207e04b209a664cdd86f4dfd7",
              "IPY_MODEL_407d79bfe1cd4782aab16e0e5fdefee8",
              "IPY_MODEL_178ffd10380940f1a943f79ea9de7f92"
            ],
            "layout": "IPY_MODEL_2a60b4f0ebac4829b0273798ab40e940"
          }
        },
        "a3e0192207e04b209a664cdd86f4dfd7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b7165540b438429aab942db315e68cc9",
            "placeholder": "​",
            "style": "IPY_MODEL_56b63656744c433fa861082e03e32dda",
            "value": "Loading checkpoint shards: 100%"
          }
        },
        "407d79bfe1cd4782aab16e0e5fdefee8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b3c4b1f3578346799154aed528166f6c",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_290b2567bab34bbaa4372fadc8945861",
            "value": 2
          }
        },
        "178ffd10380940f1a943f79ea9de7f92": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a772a77f604940c48db48f1c49a44ec0",
            "placeholder": "​",
            "style": "IPY_MODEL_9bf2f33b5b774827b38117eaace3f890",
            "value": " 2/2 [01:03&lt;00:00, 29.05s/it]"
          }
        },
        "2a60b4f0ebac4829b0273798ab40e940": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b7165540b438429aab942db315e68cc9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "56b63656744c433fa861082e03e32dda": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b3c4b1f3578346799154aed528166f6c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "290b2567bab34bbaa4372fadc8945861": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "a772a77f604940c48db48f1c49a44ec0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9bf2f33b5b774827b38117eaace3f890": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "4875edb91c9643ad9dcef7d6cf85fcaa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_d2885421bec64972b7c95599eebd211c",
              "IPY_MODEL_173474c1a07646d690ff8bf3f0727cc3",
              "IPY_MODEL_94cd0be1871a4b09918344dd61d7e0ff"
            ],
            "layout": "IPY_MODEL_8a2dfaa1ea0c4492ae7b7dd39a64d661"
          }
        },
        "d2885421bec64972b7c95599eebd211c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fd0524ed03e341d4b5b92a6096be9698",
            "placeholder": "​",
            "style": "IPY_MODEL_05a2dcc1f1ee4e2c9698d0bee576c5f7",
            "value": "Downloading (…)neration_config.json: 100%"
          }
        },
        "173474c1a07646d690ff8bf3f0727cc3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_73b2d743d3444b8ea746bbd5b8d23d5a",
            "max": 188,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_55c4441adc5744c99d57cb571563f082",
            "value": 188
          }
        },
        "94cd0be1871a4b09918344dd61d7e0ff": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9f172f8146e945d18d4b544b0e5d99a2",
            "placeholder": "​",
            "style": "IPY_MODEL_885c4f614b124d46a87707432479388c",
            "value": " 188/188 [00:00&lt;00:00, 11.0kB/s]"
          }
        },
        "8a2dfaa1ea0c4492ae7b7dd39a64d661": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fd0524ed03e341d4b5b92a6096be9698": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "05a2dcc1f1ee4e2c9698d0bee576c5f7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "73b2d743d3444b8ea746bbd5b8d23d5a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "55c4441adc5744c99d57cb571563f082": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "9f172f8146e945d18d4b544b0e5d99a2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "885c4f614b124d46a87707432479388c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/mflechl/colab-genai-simple/blob/main/finetuning_llama2_7b_1gpu.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "adapted from https://colab.research.google.com/drive/1RojNBnLIsmm98sSNYAPC6d5WVVFX4tE9?usp=sharing"
      ],
      "metadata": {
        "id": "epqbwj-dMWIj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Ludwig + DeepLearning.ai: Efficient Fine-Tuning for Llama2-7b on a Single GPU** 🙌\n",
        "\n",
        "Let's explore how to fine-tune an LLM on a single commodity GPU with [Ludwig](https://ludwig.ai/latest/), an open-source package that empowers you to effortlessly build and train machine learning models like LLMs, neural networks and tree based models through declarative config files.\n",
        "\n",
        "In this notebook, we'll show an example of how to fine-tune Llama-2-7b to generate code using the CodeAlpaca dataset.\n",
        "\n",
        "By the end of this example, you will have gained a comprehensive understanding of the following key aspects:\n",
        "\n",
        "1. **Ludwig**: An intuitive toolkit that simplifies fine-tuning for open-source Language Model Models (LLMs).\n",
        "2. **Exploring the base model with prompts**: Dive into the intricacies of prompts and prompt templates, unlocking new dimensions in LLM interaction.\n",
        "3. **Fine-Tuning Large Language Models**: Navigate the world of model fine-tuning optimizations for getting the most out of a single memory-contrained GPU, including: LoRA and 4-bit quantization.\n",
        "\n"
      ],
      "metadata": {
        "id": "xb1aLHZRFrwA"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Install Ludwig and Ludwig's LLM related dependencies.**"
      ],
      "metadata": {
        "id": "wXMVoEFkPXJJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Install Ludwig from the latest release"
      ],
      "metadata": {
        "id": "PiZYaiRufHfh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip uninstall -y tensorflow --quiet\n",
        "!pip install ludwig\n",
        "!pip install ludwig[llm]"
      ],
      "metadata": {
        "id": "jvL1cL6wYtWz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a3aca3ae-7b18-47d9-c38e-4cb2d84c5e1f"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[33mWARNING: Skipping tensorflow as it is not installed.\u001b[0m\u001b[33m\n",
            "\u001b[0mRequirement already satisfied: ludwig in /usr/local/lib/python3.10/dist-packages (0.8.3)\n",
            "Requirement already satisfied: Cython>=0.25 in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.0.2)\n",
            "Requirement already satisfied: h5py!=3.0.0,>=2.6 in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.9.0)\n",
            "Requirement already satisfied: numpy>=1.15 in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.23.5)\n",
            "Requirement already satisfied: pandas!=1.1.5,>=1.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.5.3)\n",
            "Requirement already satisfied: scipy>=0.18 in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.11.2)\n",
            "Requirement already satisfied: tabulate>=0.7 in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.9.0)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.2.2)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from ludwig) (4.66.1)\n",
            "Requirement already satisfied: torch>=1.13.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (2.0.1+cu118)\n",
            "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (from ludwig) (2.0.2+cu118)\n",
            "Requirement already satisfied: torchtext in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.15.2)\n",
            "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.15.2+cu118)\n",
            "Requirement already satisfied: pydantic<2.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.10.12)\n",
            "Requirement already satisfied: transformers in /usr/local/lib/python3.10/dist-packages (from ludwig) (4.33.2)\n",
            "Requirement already satisfied: tokenizers>=0.13.3 in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.13.3)\n",
            "Requirement already satisfied: spacy>=2.3 in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.6.1)\n",
            "Requirement already satisfied: PyYAML!=5.4.*,<6.0.1,>=3.12 in /usr/local/lib/python3.10/dist-packages (from ludwig) (6.0)\n",
            "Requirement already satisfied: absl-py in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.4.0)\n",
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.5.16)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from ludwig) (2.31.0)\n",
            "Requirement already satisfied: fsspec[http] in /usr/local/lib/python3.10/dist-packages (from ludwig) (2023.6.0)\n",
            "Requirement already satisfied: dataclasses-json in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.6.0)\n",
            "Requirement already satisfied: jsonschema<4.7,>=4.5.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (4.6.2)\n",
            "Requirement already satisfied: marshmallow in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.20.1)\n",
            "Requirement already satisfied: marshmallow-jsonschema in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.13.0)\n",
            "Requirement already satisfied: marshmallow-dataclass==8.5.4 in /usr/local/lib/python3.10/dist-packages (from ludwig) (8.5.4)\n",
            "Requirement already satisfied: tensorboard in /usr/local/lib/python3.10/dist-packages (from ludwig) (2.13.0)\n",
            "Requirement already satisfied: nltk in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.8.1)\n",
            "Requirement already satisfied: torchmetrics<=0.11.4,>=0.11.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.11.4)\n",
            "Requirement already satisfied: torchinfo in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.8.0)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.12.2)\n",
            "Requirement already satisfied: psutil==5.9.4 in /usr/local/lib/python3.10/dist-packages (from ludwig) (5.9.4)\n",
            "Requirement already satisfied: protobuf==3.20.3 in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.20.3)\n",
            "Requirement already satisfied: py-cpuinfo==9.0.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (9.0.0)\n",
            "Requirement already satisfied: gpustat in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.1.1)\n",
            "Requirement already satisfied: rich~=12.4.4 in /usr/local/lib/python3.10/dist-packages (from ludwig) (12.4.4)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from ludwig) (23.1)\n",
            "Requirement already satisfied: retry in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.9.2)\n",
            "Requirement already satisfied: sacremoses in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.0.53)\n",
            "Requirement already satisfied: sentencepiece in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.1.99)\n",
            "Requirement already satisfied: getdaft in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.1.17)\n",
            "Requirement already satisfied: bitsandbytes<0.41.0 in /usr/local/lib/python3.10/dist-packages (from ludwig) (0.40.2)\n",
            "Requirement already satisfied: xlwt in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.3.0)\n",
            "Requirement already satisfied: xlrd>=2.0.1 in /usr/local/lib/python3.10/dist-packages (from ludwig) (2.0.1)\n",
            "Requirement already satisfied: xlsxwriter>=1.4.3 in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.1.3)\n",
            "Requirement already satisfied: openpyxl>=3.0.7 in /usr/local/lib/python3.10/dist-packages (from ludwig) (3.1.2)\n",
            "Requirement already satisfied: pyxlsb>=1.0.8 in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.0.10)\n",
            "Requirement already satisfied: pyarrow in /usr/local/lib/python3.10/dist-packages (from ludwig) (9.0.0)\n",
            "Requirement already satisfied: lxml in /usr/local/lib/python3.10/dist-packages (from ludwig) (4.9.3)\n",
            "Requirement already satisfied: html5lib in /usr/local/lib/python3.10/dist-packages (from ludwig) (1.1)\n",
            "Requirement already satisfied: typing-inspect>=0.7.1 in /usr/local/lib/python3.10/dist-packages (from marshmallow-dataclass==8.5.4->ludwig) (0.9.0)\n",
            "Requirement already satisfied: attrs>=17.4.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema<4.7,>=4.5.0->ludwig) (23.1.0)\n",
            "Requirement already satisfied: pyrsistent!=0.17.0,!=0.17.1,!=0.17.2,>=0.14.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema<4.7,>=4.5.0->ludwig) (0.19.3)\n",
            "Requirement already satisfied: et-xmlfile in /usr/local/lib/python3.10/dist-packages (from openpyxl>=3.0.7->ludwig) (1.1.0)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas!=1.1.5,>=1.0->ludwig) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas!=1.1.5,>=1.0->ludwig) (2023.3.post1)\n",
            "Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<2.0->ludwig) (4.5.0)\n",
            "Requirement already satisfied: commonmark<0.10.0,>=0.9.0 in /usr/local/lib/python3.10/dist-packages (from rich~=12.4.4->ludwig) (0.9.1)\n",
            "Requirement already satisfied: pygments<3.0.0,>=2.6.0 in /usr/local/lib/python3.10/dist-packages (from rich~=12.4.4->ludwig) (2.16.1)\n",
            "Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.11 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (3.0.12)\n",
            "Requirement already satisfied: spacy-loggers<2.0.0,>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (1.0.4)\n",
            "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (1.0.9)\n",
            "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (2.0.7)\n",
            "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (3.0.8)\n",
            "Requirement already satisfied: thinc<8.2.0,>=8.1.8 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (8.1.12)\n",
            "Requirement already satisfied: wasabi<1.2.0,>=0.9.1 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (1.1.2)\n",
            "Requirement already satisfied: srsly<3.0.0,>=2.4.3 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (2.4.7)\n",
            "Requirement already satisfied: catalogue<2.1.0,>=2.0.6 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (2.0.9)\n",
            "Requirement already satisfied: typer<0.10.0,>=0.3.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (0.9.0)\n",
            "Requirement already satisfied: pathy>=0.10.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (0.10.2)\n",
            "Requirement already satisfied: smart-open<7.0.0,>=5.2.1 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (6.4.0)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (3.1.2)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (67.7.2)\n",
            "Requirement already satisfied: langcodes<4.0.0,>=3.2.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig) (3.3.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig) (3.2.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig) (3.4)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig) (2.0.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig) (2023.7.22)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch>=1.13.0->ludwig) (1.12)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.13.0->ludwig) (3.1)\n",
            "Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.13.0->ludwig) (2.0.0)\n",
            "Requirement already satisfied: cmake in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.13.0->ludwig) (3.27.4.1)\n",
            "Requirement already satisfied: lit in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.13.0->ludwig) (16.0.6)\n",
            "Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in /usr/local/lib/python3.10/dist-packages (from fsspec[http]->ludwig) (3.8.5)\n",
            "Requirement already satisfied: loguru in /usr/local/lib/python3.10/dist-packages (from getdaft->ludwig) (0.7.2)\n",
            "Requirement already satisfied: nvidia-ml-py>=11.450.129 in /usr/local/lib/python3.10/dist-packages (from gpustat->ludwig) (12.535.108)\n",
            "Requirement already satisfied: blessed>=1.17.1 in /usr/local/lib/python3.10/dist-packages (from gpustat->ludwig) (1.20.0)\n",
            "Requirement already satisfied: six>=1.9 in /usr/local/lib/python3.10/dist-packages (from html5lib->ludwig) (1.16.0)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from html5lib->ludwig) (0.5.1)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.10/dist-packages (from kaggle->ludwig) (8.0.1)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from kaggle->ludwig) (6.0.0)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.10/dist-packages (from nltk->ludwig) (8.1.7)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.10/dist-packages (from nltk->ludwig) (1.3.2)\n",
            "Requirement already satisfied: regex>=2021.8.3 in /usr/local/lib/python3.10/dist-packages (from nltk->ludwig) (2023.6.3)\n",
            "Requirement already satisfied: decorator>=3.4.2 in /usr/local/lib/python3.10/dist-packages (from retry->ludwig) (4.4.2)\n",
            "Requirement already satisfied: py<2.0.0,>=1.4.26 in /usr/local/lib/python3.10/dist-packages (from retry->ludwig) (1.11.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->ludwig) (3.2.0)\n",
            "Requirement already satisfied: grpcio>=1.48.2 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (1.57.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (2.17.3)\n",
            "Requirement already satisfied: google-auth-oauthlib<1.1,>=0.5 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (1.0.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (3.4.4)\n",
            "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (0.7.1)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (2.3.7)\n",
            "Requirement already satisfied: wheel>=0.26 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig) (0.41.2)\n",
            "Requirement already satisfied: torchdata==0.6.1 in /usr/local/lib/python3.10/dist-packages (from torchtext->ludwig) (0.6.1)\n",
            "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision->ludwig) (9.4.0)\n",
            "Requirement already satisfied: huggingface-hub<1.0,>=0.15.1 in /usr/local/lib/python3.10/dist-packages (from transformers->ludwig) (0.16.4)\n",
            "Requirement already satisfied: safetensors>=0.3.1 in /usr/local/lib/python3.10/dist-packages (from transformers->ludwig) (0.3.3)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig) (6.0.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig) (4.0.3)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig) (1.9.2)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig) (1.4.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig) (1.3.1)\n",
            "Requirement already satisfied: wcwidth>=0.1.4 in /usr/local/lib/python3.10/dist-packages (from blessed>=1.17.1->gpustat->ludwig) (0.2.6)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard->ludwig) (5.3.1)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard->ludwig) (0.3.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard->ludwig) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<1.1,>=0.5->tensorboard->ludwig) (1.3.1)\n",
            "Requirement already satisfied: blis<0.8.0,>=0.7.8 in /usr/local/lib/python3.10/dist-packages (from thinc<8.2.0,>=8.1.8->spacy>=2.3->ludwig) (0.7.10)\n",
            "Requirement already satisfied: confection<1.0.0,>=0.0.1 in /usr/local/lib/python3.10/dist-packages (from thinc<8.2.0,>=8.1.8->spacy>=2.3->ludwig) (0.1.2)\n",
            "Requirement already satisfied: mypy-extensions>=0.3.0 in /usr/local/lib/python3.10/dist-packages (from typing-inspect>=0.7.1->marshmallow-dataclass==8.5.4->ludwig) (1.0.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard->ludwig) (2.1.3)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.10/dist-packages (from python-slugify->kaggle->ludwig) (1.3)\n",
            "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch>=1.13.0->ludwig) (1.3.0)\n",
            "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard->ludwig) (0.5.0)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<1.1,>=0.5->tensorboard->ludwig) (3.2.2)\n",
            "Requirement already satisfied: ludwig[llm] in /usr/local/lib/python3.10/dist-packages (0.8.3)\n",
            "Requirement already satisfied: Cython>=0.25 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.0.2)\n",
            "Requirement already satisfied: h5py!=3.0.0,>=2.6 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.9.0)\n",
            "Requirement already satisfied: numpy>=1.15 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.23.5)\n",
            "Requirement already satisfied: pandas!=1.1.5,>=1.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.5.3)\n",
            "Requirement already satisfied: scipy>=0.18 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.11.2)\n",
            "Requirement already satisfied: tabulate>=0.7 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.9.0)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.2.2)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (4.66.1)\n",
            "Requirement already satisfied: torch>=1.13.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2.0.1+cu118)\n",
            "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2.0.2+cu118)\n",
            "Requirement already satisfied: torchtext in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.15.2)\n",
            "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.15.2+cu118)\n",
            "Requirement already satisfied: pydantic<2.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.10.12)\n",
            "Requirement already satisfied: transformers in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (4.33.2)\n",
            "Requirement already satisfied: tokenizers>=0.13.3 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.13.3)\n",
            "Requirement already satisfied: spacy>=2.3 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.6.1)\n",
            "Requirement already satisfied: PyYAML!=5.4.*,<6.0.1,>=3.12 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (6.0)\n",
            "Requirement already satisfied: absl-py in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.4.0)\n",
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.5.16)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2.31.0)\n",
            "Requirement already satisfied: fsspec[http] in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2023.6.0)\n",
            "Requirement already satisfied: dataclasses-json in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.6.0)\n",
            "Requirement already satisfied: jsonschema<4.7,>=4.5.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (4.6.2)\n",
            "Requirement already satisfied: marshmallow in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.20.1)\n",
            "Requirement already satisfied: marshmallow-jsonschema in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.13.0)\n",
            "Requirement already satisfied: marshmallow-dataclass==8.5.4 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (8.5.4)\n",
            "Requirement already satisfied: tensorboard in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2.13.0)\n",
            "Requirement already satisfied: nltk in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.8.1)\n",
            "Requirement already satisfied: torchmetrics<=0.11.4,>=0.11.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.11.4)\n",
            "Requirement already satisfied: torchinfo in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.8.0)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.12.2)\n",
            "Requirement already satisfied: psutil==5.9.4 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (5.9.4)\n",
            "Requirement already satisfied: protobuf==3.20.3 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.20.3)\n",
            "Requirement already satisfied: py-cpuinfo==9.0.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (9.0.0)\n",
            "Requirement already satisfied: gpustat in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.1.1)\n",
            "Requirement already satisfied: rich~=12.4.4 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (12.4.4)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (23.1)\n",
            "Requirement already satisfied: retry in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.9.2)\n",
            "Requirement already satisfied: sacremoses in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.0.53)\n",
            "Requirement already satisfied: sentencepiece in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.1.99)\n",
            "Requirement already satisfied: getdaft in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.1.17)\n",
            "Requirement already satisfied: bitsandbytes<0.41.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.40.2)\n",
            "Requirement already satisfied: xlwt in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.3.0)\n",
            "Requirement already satisfied: xlrd>=2.0.1 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2.0.1)\n",
            "Requirement already satisfied: xlsxwriter>=1.4.3 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.1.3)\n",
            "Requirement already satisfied: openpyxl>=3.0.7 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (3.1.2)\n",
            "Requirement already satisfied: pyxlsb>=1.0.8 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.0.10)\n",
            "Requirement already satisfied: pyarrow in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (9.0.0)\n",
            "Requirement already satisfied: lxml in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (4.9.3)\n",
            "Requirement already satisfied: html5lib in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.1)\n",
            "Requirement already satisfied: sentence-transformers in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (2.2.2)\n",
            "Requirement already satisfied: faiss-cpu in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (1.7.4)\n",
            "Requirement already satisfied: accelerate in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.23.0)\n",
            "Requirement already satisfied: loralib in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.1.2)\n",
            "Requirement already satisfied: peft>=0.4.0 in /usr/local/lib/python3.10/dist-packages (from ludwig[llm]) (0.5.0)\n",
            "Requirement already satisfied: typing-inspect>=0.7.1 in /usr/local/lib/python3.10/dist-packages (from marshmallow-dataclass==8.5.4->ludwig[llm]) (0.9.0)\n",
            "Requirement already satisfied: attrs>=17.4.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema<4.7,>=4.5.0->ludwig[llm]) (23.1.0)\n",
            "Requirement already satisfied: pyrsistent!=0.17.0,!=0.17.1,!=0.17.2,>=0.14.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema<4.7,>=4.5.0->ludwig[llm]) (0.19.3)\n",
            "Requirement already satisfied: et-xmlfile in /usr/local/lib/python3.10/dist-packages (from openpyxl>=3.0.7->ludwig[llm]) (1.1.0)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas!=1.1.5,>=1.0->ludwig[llm]) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas!=1.1.5,>=1.0->ludwig[llm]) (2023.3.post1)\n",
            "Requirement already satisfied: safetensors in /usr/local/lib/python3.10/dist-packages (from peft>=0.4.0->ludwig[llm]) (0.3.3)\n",
            "Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<2.0->ludwig[llm]) (4.5.0)\n",
            "Requirement already satisfied: commonmark<0.10.0,>=0.9.0 in /usr/local/lib/python3.10/dist-packages (from rich~=12.4.4->ludwig[llm]) (0.9.1)\n",
            "Requirement already satisfied: pygments<3.0.0,>=2.6.0 in /usr/local/lib/python3.10/dist-packages (from rich~=12.4.4->ludwig[llm]) (2.16.1)\n",
            "Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.11 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (3.0.12)\n",
            "Requirement already satisfied: spacy-loggers<2.0.0,>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (1.0.4)\n",
            "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (1.0.9)\n",
            "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (2.0.7)\n",
            "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (3.0.8)\n",
            "Requirement already satisfied: thinc<8.2.0,>=8.1.8 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (8.1.12)\n",
            "Requirement already satisfied: wasabi<1.2.0,>=0.9.1 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (1.1.2)\n",
            "Requirement already satisfied: srsly<3.0.0,>=2.4.3 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (2.4.7)\n",
            "Requirement already satisfied: catalogue<2.1.0,>=2.0.6 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (2.0.9)\n",
            "Requirement already satisfied: typer<0.10.0,>=0.3.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (0.9.0)\n",
            "Requirement already satisfied: pathy>=0.10.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (0.10.2)\n",
            "Requirement already satisfied: smart-open<7.0.0,>=5.2.1 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (6.4.0)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (3.1.2)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (67.7.2)\n",
            "Requirement already satisfied: langcodes<4.0.0,>=3.2.0 in /usr/local/lib/python3.10/dist-packages (from spacy>=2.3->ludwig[llm]) (3.3.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig[llm]) (3.2.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig[llm]) (3.4)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig[llm]) (2.0.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->ludwig[llm]) (2023.7.22)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch>=1.13.0->ludwig[llm]) (1.12)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.13.0->ludwig[llm]) (3.1)\n",
            "Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.13.0->ludwig[llm]) (2.0.0)\n",
            "Requirement already satisfied: cmake in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.13.0->ludwig[llm]) (3.27.4.1)\n",
            "Requirement already satisfied: lit in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.13.0->ludwig[llm]) (16.0.6)\n",
            "Requirement already satisfied: huggingface-hub in /usr/local/lib/python3.10/dist-packages (from accelerate->ludwig[llm]) (0.16.4)\n",
            "Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in /usr/local/lib/python3.10/dist-packages (from fsspec[http]->ludwig[llm]) (3.8.5)\n",
            "Requirement already satisfied: loguru in /usr/local/lib/python3.10/dist-packages (from getdaft->ludwig[llm]) (0.7.2)\n",
            "Requirement already satisfied: nvidia-ml-py>=11.450.129 in /usr/local/lib/python3.10/dist-packages (from gpustat->ludwig[llm]) (12.535.108)\n",
            "Requirement already satisfied: blessed>=1.17.1 in /usr/local/lib/python3.10/dist-packages (from gpustat->ludwig[llm]) (1.20.0)\n",
            "Requirement already satisfied: six>=1.9 in /usr/local/lib/python3.10/dist-packages (from html5lib->ludwig[llm]) (1.16.0)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from html5lib->ludwig[llm]) (0.5.1)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.10/dist-packages (from kaggle->ludwig[llm]) (8.0.1)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from kaggle->ludwig[llm]) (6.0.0)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.10/dist-packages (from nltk->ludwig[llm]) (8.1.7)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.10/dist-packages (from nltk->ludwig[llm]) (1.3.2)\n",
            "Requirement already satisfied: regex>=2021.8.3 in /usr/local/lib/python3.10/dist-packages (from nltk->ludwig[llm]) (2023.6.3)\n",
            "Requirement already satisfied: decorator>=3.4.2 in /usr/local/lib/python3.10/dist-packages (from retry->ludwig[llm]) (4.4.2)\n",
            "Requirement already satisfied: py<2.0.0,>=1.4.26 in /usr/local/lib/python3.10/dist-packages (from retry->ludwig[llm]) (1.11.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->ludwig[llm]) (3.2.0)\n",
            "Requirement already satisfied: grpcio>=1.48.2 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (1.57.0)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (2.17.3)\n",
            "Requirement already satisfied: google-auth-oauthlib<1.1,>=0.5 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (1.0.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (3.4.4)\n",
            "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (0.7.1)\n",
            "Requirement already satisfied: werkzeug>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (2.3.7)\n",
            "Requirement already satisfied: wheel>=0.26 in /usr/local/lib/python3.10/dist-packages (from tensorboard->ludwig[llm]) (0.41.2)\n",
            "Requirement already satisfied: torchdata==0.6.1 in /usr/local/lib/python3.10/dist-packages (from torchtext->ludwig[llm]) (0.6.1)\n",
            "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision->ludwig[llm]) (9.4.0)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig[llm]) (6.0.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig[llm]) (4.0.3)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig[llm]) (1.9.2)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig[llm]) (1.4.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]->ludwig[llm]) (1.3.1)\n",
            "Requirement already satisfied: wcwidth>=0.1.4 in /usr/local/lib/python3.10/dist-packages (from blessed>=1.17.1->gpustat->ludwig[llm]) (0.2.6)\n",
            "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard->ludwig[llm]) (5.3.1)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard->ludwig[llm]) (0.3.0)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.10/dist-packages (from google-auth<3,>=1.6.3->tensorboard->ludwig[llm]) (4.9)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from google-auth-oauthlib<1.1,>=0.5->tensorboard->ludwig[llm]) (1.3.1)\n",
            "Requirement already satisfied: blis<0.8.0,>=0.7.8 in /usr/local/lib/python3.10/dist-packages (from thinc<8.2.0,>=8.1.8->spacy>=2.3->ludwig[llm]) (0.7.10)\n",
            "Requirement already satisfied: confection<1.0.0,>=0.0.1 in /usr/local/lib/python3.10/dist-packages (from thinc<8.2.0,>=8.1.8->spacy>=2.3->ludwig[llm]) (0.1.2)\n",
            "Requirement already satisfied: mypy-extensions>=0.3.0 in /usr/local/lib/python3.10/dist-packages (from typing-inspect>=0.7.1->marshmallow-dataclass==8.5.4->ludwig[llm]) (1.0.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.1.1 in /usr/local/lib/python3.10/dist-packages (from werkzeug>=1.0.1->tensorboard->ludwig[llm]) (2.1.3)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.10/dist-packages (from python-slugify->kaggle->ludwig[llm]) (1.3)\n",
            "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch>=1.13.0->ludwig[llm]) (1.3.0)\n",
            "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /usr/local/lib/python3.10/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard->ludwig[llm]) (0.5.0)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.10/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<1.1,>=0.5->tensorboard->ludwig[llm]) (3.2.2)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Install Ludwig from Ludwig master"
      ],
      "metadata": {
        "id": "C7eAUs6efGH5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# !pip uninstall -y tensorflow --quiet\n",
        "# !pip install git+https://github.com/ludwig-ai/ludwig.git@master --quiet\n",
        "# !pip install \"git+https://github.com/ludwig-ai/ludwig.git@master#egg=ludwig[llm]\" --quiet"
      ],
      "metadata": {
        "id": "yfYXyUUvM-R6"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Enable text wrapping so we don't have to scroll horizontally and create a function to flush CUDA cache."
      ],
      "metadata": {
        "id": "A2SRHcKAJYuu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from IPython.display import HTML, display\n",
        "\n",
        "def set_css():\n",
        "  display(HTML('''\n",
        "  <style>\n",
        "    pre {\n",
        "        white-space: pre-wrap;\n",
        "    }\n",
        "  </style>\n",
        "  '''))\n",
        "\n",
        "get_ipython().events.register('pre_run_cell', set_css)\n",
        "\n",
        "def clear_cache():\n",
        "  if torch.cuda.is_available():\n",
        "    torch.cuda.empty_cache()"
      ],
      "metadata": {
        "id": "Ht4eVWxB13QL"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Setup Your HuggingFace Token** 🤗\n",
        "\n",
        "We'll be exploring Llama-2 today, which a model released by Meta. However, the model is not openly-accessible and requires requesting for access (assigned to your HuggingFace token).\n",
        "\n",
        "Obtain a [HuggingFace API Token](https://huggingface.co/settings/tokens) and request access to [Llama2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) before proceeding. You may need to signup on HuggingFace if you don't aleady have an account: https://huggingface.co/join\n",
        "\n",
        "Incase you haven't been given access to Llama-2-7b, that is alright. We can just use Llama-1 for the rest of this example: [huggyllama/llama-7b](https://huggingface.co/huggyllama/llama-7b)."
      ],
      "metadata": {
        "id": "SUkPQ9OLP0Po"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import getpass\n",
        "import locale; locale.getpreferredencoding = lambda: \"UTF-8\"\n",
        "import logging\n",
        "import os\n",
        "import torch\n",
        "import yaml\n",
        "\n",
        "from ludwig.api import LudwigModel\n",
        "\n",
        "\n",
        "os.environ[\"HUGGING_FACE_HUB_TOKEN\"] = getpass.getpass(\"Token:\")\n",
        "assert os.environ[\"HUGGING_FACE_HUB_TOKEN\"]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "id": "nXOIxmfbQHSz",
        "outputId": "eccd6843-299d-409e-872c-8c7309b55da2"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Token:··········\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Import The Code Generation Dataset** 📋\n",
        "\n"
      ],
      "metadata": {
        "id": "sV0D2rjRbk4z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "cAsqyCDX-_NA",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "ae83e73d-1868-4afc-c146-90357293c68e"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# get from here:\n",
        "# https://huggingface.co/datasets/ArtifactAI/arxiv-math-instruct-50k/tree/main\n",
        "# https://huggingface.co/datasets/ArtifactAI/arxiv-physics-instruct-tune-30k/tree/main\n",
        "\n",
        "!ls -l /content/drive/MyDrive/llama2-ft/arxiv_physics_instruct_30k.jsonl /content/drive/MyDrive/llama2-ft/arxiv_math_instruct_50k.jsonl"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 86
        },
        "id": "0hXIXO3f1yUY",
        "outputId": "26d30500-f370-49d0-d6c5-fd6b84c4e79e"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "-rw------- 1 root root 36635781 Sep 17 15:53 /content/drive/MyDrive/llama2-ft/arxiv_math_instruct_50k.jsonl\n",
            "-rw------- 1 root root 21829218 Sep 17 15:52 /content/drive/MyDrive/llama2-ft/arxiv_physics_instruct_30k.jsonl\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import data_table; data_table.enable_dataframe_formatter()\n",
        "import numpy as np; np.random.seed(123)\n",
        "import pandas as pd\n",
        "\n",
        "import json\n",
        "with open('/content/drive/MyDrive/llama2-ft/arxiv_physics_instruct_30k.jsonl') as f1:\n",
        "    data1 = [json.loads(line) for line in f1]\n",
        "with open('/content/drive/MyDrive/llama2-ft/arxiv_math_instruct_50k.jsonl') as f2:\n",
        "    data2 = [json.loads(line) for line in f2]\n",
        "\n",
        "df1 = pd.DataFrame(data1)\n",
        "df2 = pd.DataFrame(data2)\n",
        "df = pd.concat([df1, df2])\n",
        "main_df = df.sample(frac=1, random_state=42)\n",
        "main_df.reset_index(drop=True, inplace=True)\n",
        "# We're going to create a new column called `split` where:\n",
        "# 90% will be assigned a value of 0 -> train set\n",
        "# 5% will be assigned a value of 1 -> validation set\n",
        "# 5% will be assigned a value of 2 -> test set\n",
        "# Calculate the number of rows for each split value\n",
        "total_rows = len(main_df)\n",
        "split_0_count = int(total_rows * 0.9)\n",
        "split_1_count = int(total_rows * 0.05)\n",
        "split_2_count = total_rows - split_0_count - split_1_count\n",
        "\n",
        "# Create an array with split values based on the counts\n",
        "split_values = np.concatenate([\n",
        "    np.zeros(split_0_count),\n",
        "    np.ones(split_1_count),\n",
        "    np.full(split_2_count, 2)\n",
        "])\n",
        "\n",
        "# Shuffle the array to ensure randomness\n",
        "np.random.shuffle(split_values)\n",
        "\n",
        "# Add the 'split' column to the DataFrame\n",
        "main_df['split'] = split_values\n",
        "main_df['split'] = main_df['split'].astype(int)\n",
        "\n",
        "# For this webinar, we will just use 1000 rows of this dataset.\n",
        "main_df = main_df.head(n=500)"
      ],
      "metadata": {
        "id": "m8FXsCwYbkHh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "0b1d774e-abf2-443c-ca8e-abcc2d350508"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(main_df)"
      ],
      "metadata": {
        "id": "bvfE3EZ677WT",
        "outputId": "24261482-1897-47e8-dcc1-b5c015984f90",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "500"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "main_df.head(n=10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1212
        },
        "id": "Mx3JQMpmb1KQ",
        "outputId": "c69a85e2-2f3a-4b2c-cba2-00f4d3bd4958"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                            question  \\\n",
              "0  Which homology determines fibers over the circle?   \n",
              "1         At what magnitude are debris disks formed?   \n",
              "2                 How do you measure casimir forces?   \n",
              "3  What are the three basic features of the unive...   \n",
              "4  What type of field is found on jets with helic...   \n",
              "5            What is the quantum potential madelung?   \n",
              "6  What is the mathematical term for almost ordin...   \n",
              "7                   How does dsph evolute in hordes?   \n",
              "8  What is the second harmonic susceptibility ten...   \n",
              "9  What is the cross section of central diffracti...   \n",
              "\n",
              "                                              answer  split  \n",
              "0  The homology that determines fibers over the c...      0  \n",
              "1  Debris disks are typically formed around stars...      0  \n",
              "2  Casimir forces are measured using a specialize...      1  \n",
              "3  As an AI language model, there is no scientifi...      0  \n",
              "4  On jets with helical fields, the magnetic fiel...      0  \n",
              "5  The Madelung or quantum potential is a concept...      0  \n",
              "6  The mathematical term for almost ordinary Galo...      0  \n",
              "7  Dwarf spheroidal galaxies (dSph) in hordes evo...      0  \n",
              "8  The second harmonic susceptibility tensor is a...      0  \n",
              "9  The cross section of central diffractive Higgs...      0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-bf5ab77a-264d-4c02-bfdc-0ddd6f3eaffe\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>question</th>\n",
              "      <th>answer</th>\n",
              "      <th>split</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Which homology determines fibers over the circle?</td>\n",
              "      <td>The homology that determines fibers over the c...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>At what magnitude are debris disks formed?</td>\n",
              "      <td>Debris disks are typically formed around stars...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>How do you measure casimir forces?</td>\n",
              "      <td>Casimir forces are measured using a specialize...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>What are the three basic features of the unive...</td>\n",
              "      <td>As an AI language model, there is no scientifi...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>What type of field is found on jets with helic...</td>\n",
              "      <td>On jets with helical fields, the magnetic fiel...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>What is the quantum potential madelung?</td>\n",
              "      <td>The Madelung or quantum potential is a concept...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>What is the mathematical term for almost ordin...</td>\n",
              "      <td>The mathematical term for almost ordinary Galo...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>How does dsph evolute in hordes?</td>\n",
              "      <td>Dwarf spheroidal galaxies (dSph) in hordes evo...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>What is the second harmonic susceptibility ten...</td>\n",
              "      <td>The second harmonic susceptibility tensor is a...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>What is the cross section of central diffracti...</td>\n",
              "      <td>The cross section of central diffractive Higgs...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bf5ab77a-264d-4c02-bfdc-0ddd6f3eaffe')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-bf5ab77a-264d-4c02-bfdc-0ddd6f3eaffe button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-bf5ab77a-264d-4c02-bfdc-0ddd6f3eaffe');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-a1c0e10e-1301-4226-8bfc-2515a082d6d3\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-a1c0e10e-1301-4226-8bfc-2515a082d6d3')\"\n",
              "            title=\"Suggest charts.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-a1c0e10e-1301-4226-8bfc-2515a082d6d3 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.module+javascript": "\n      import \"https://ssl.gstatic.com/colaboratory/data_table/881c4a0d49046431/data_table.js\";\n\n      const table = window.createDataTable({\n        data: [[{\n            'v': 0,\n            'f': \"0\",\n        },\n\"Which homology determines fibers over the circle?\",\n\"The homology that determines fibers over the circle is the first singular homology group with integer coefficients, denoted as $H_1(X,\\\\mathbb{Z})$, where $X$ is the space whose fibers are being considered. This homology group captures the one-dimensional loops in $X$ that wrap around the circle. Specifically, if $p:E \\\\rightarrow B$ is a fiber bundle with fiber $F$, then $H_1(B,\\\\mathbb{Z})$ acts as the coefficient group for $H_1(F,\\\\mathbb{Z})$ via the induced pushforward homomorphism $p_*:H_1(F,\\\\mathbb{Z}) \\\\rightarrow H_1(B,\\\\mathbb{Z})$. In other words, $H_1(B,\\\\mathbb{Z})$ determine the ways in which the circle \\\"wraps\\\" around the fibers of the bundle.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 1,\n            'f': \"1\",\n        },\n\"At what magnitude are debris disks formed?\",\n\"Debris disks are typically formed around stars that are at or slightly older than the age of our own Sun (around 5 billion years old), when the star's planetary system has had enough time to mature and form rocky and icy bodies that can collide and create debris. These debris disks are usually detected as excess infrared radiation coming from the star, indicating the presence of dusty material.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 2,\n            'f': \"2\",\n        },\n\"How do you measure casimir forces?\",\n\"Casimir forces are measured using a specialized experimental apparatus called a Casimir force measurement device. This device typically consists of two parallel plates placed a short distance apart in a vacuum and held in place by a small spring or string. The distance between the plates is carefully controlled, and the force between the plates is measured using a variety of techniques, such as interferometry or a torsion balance. The precise measurement of Casimir forces requires the careful control of environmental factors such as temperature, humidity, and electromagnetic interference.\",\n{\n            'v': 1,\n            'f': \"1\",\n        }],\n [{\n            'v': 3,\n            'f': \"3\",\n        },\n\"What are the three basic features of the universal tachyon?\",\n\"As an AI language model, there is no scientific consensus about the existence of tachyons or their properties. Nonetheless, from a theoretical standpoint, tachyons are hypothetical particles that would travel faster than the speed of light. Their properties are still being researched and debated, and there is no definitive agreement on what their characteristics would be. However, some proposed features of a universal tachyon include:\\n\\n1. Always traveling faster than the speed of light\\n2. Having an imaginary mass, which means it has an intrinsic energy that decreases as its speed increases\\n3. Emitting Cherenkov radiation, which is a type of electromagnetic radiation that is created when charged particles travel faster than the speed of light in a medium. \\n\\nIt is important to note that these features are purely hypothetical and are still being studied and debated in the field of theoretical physics.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 4,\n            'f': \"4\",\n        },\n\"What type of field is found on jets with helical fields?\",\n\"On jets with helical fields, the magnetic field is twisted and spiraling around the jet axis. This type of magnetic field is known as a helical magnetic field.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 5,\n            'f': \"5\",\n        },\n\"What is the quantum potential madelung?\",\n\"The Madelung or quantum potential is a concept in quantum mechanics that was introduced by physicist Erwin Madelung in 1926. It is a way of describing the behavior of wave functions in terms of a classical-like field, which is determined by the probability amplitude of the wave function.\\n\\nThe quantum potential describes the interaction between particles (or systems of particles) that are described by wave functions. It arises from the wave-like nature of particles, which leads to interference effects that cannot be explained by classical mechanics.\\n\\nIn the context of the Schr\\u00f6dinger equation, the Madelung potential is given by the gradient of the phase of the wave function, and it plays an important role in the Bohmian interpretation of quantum mechanics.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 6,\n            'f': \"6\",\n        },\n\"What is the mathematical term for almost ordinary galois deformation?\",\n\"The mathematical term for almost ordinary Galois deformation is \\\"nearly ordinary Galois deformation\\\".\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 7,\n            'f': \"7\",\n        },\n\"How does dsph evolute in hordes?\",\n\"Dwarf spheroidal galaxies (dSph) in hordes evolve differently compared to those in isolation. The interaction between neighbouring galaxies causes tidal forces, which can strip gas and stars from the dSph galaxies. This can lead to a reduction in the mass and size of the dSph galaxies over time.\\n\\nAdditionally, hordes can also cause dSph galaxies to undergo more frequent and intense interactions, which can lead to bursts of star formation as gas is compressed and triggered by gravitational forces. However, this also means that the dSph galaxies are more susceptible to being disrupted or merging with other galaxies in the horde over time.\\n\\nOverall, the evolution of dSph galaxies in hordes is a complex process that depends on a range of factors such as the mass of the horde, the distance between galaxies, and the rate of interactions between galaxies.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 8,\n            'f': \"8\",\n        },\n\"What is the second harmonic susceptibility tensor?\",\n\"The second harmonic susceptibility tensor is a quantity that characterizes how efficiently a material can generate or respond to second harmonic light, which has twice the frequency of the incoming light. It is a rank-2 tensor that relates the second harmonic polarization to the incident electric field at the fundamental frequency. It is denoted by the symbol $\\\\chi^{(2)}_{ijk}$, where i, j, and k are Cartesian indices that label the directions of the electric field components. The value of the tensor elements depends on the crystal symmetry of the material and its orientation with respect to the incident beam. In nonlinear optics, the second harmonic susceptibility tensor is a key parameter for designing optical devices and studying the fundamental behavior of light-matter interactions.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }],\n [{\n            'v': 9,\n            'f': \"9\",\n        },\n\"What is the cross section of central diffractive higgs production?\",\n\"The cross section of central diffractive Higgs production is a measure of the rate at which this process occurs in a particular experiment or setting. This quantity is determined by a variety of factors including the energy and geometry of the colliding particles, the properties of the Higgs boson, and various other experimental conditions. The cross section is typically measured in units of area, such as barns or femtobarns, and is an important parameter for understanding the physics of diffractive processes and the properties of the Higgs boson.\",\n{\n            'v': 0,\n            'f': \"0\",\n        }]],\n        columns: [[\"number\", \"index\"], [\"string\", \"question\"], [\"string\", \"answer\"], [\"number\", \"split\"]],\n        columnOptions: [{\"width\": \"1px\", \"className\": \"index_column\"}],\n        rowsPerPage: 25,\n        helpUrl: \"https://colab.research.google.com/notebooks/data_table.ipynb\",\n        suppressOutputScrolling: true,\n        minimumWidth: undefined,\n      });\n\n      function appendQuickchartButton(parentElement) {\n        let quickchartButtonContainerElement = document.createElement('div');\n        quickchartButtonContainerElement.innerHTML = `\n<div id=\"df-a5b2c845-4f6e-4022-b145-611c07499cb9\">\n  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-a5b2c845-4f6e-4022-b145-611c07499cb9')\"\n            title=\"Suggest charts.\"\n            style=\"display:none;\">\n    \n<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n     width=\"24px\">\n    <g>\n        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n    </g>\n</svg>\n  </button>\n  \n<style>\n  .colab-df-quickchart {\n      --bg-color: #E8F0FE;\n      --fill-color: #1967D2;\n      --hover-bg-color: #E2EBFA;\n      --hover-fill-color: #174EA6;\n      --disabled-fill-color: #AAA;\n      --disabled-bg-color: #DDD;\n  }\n\n  [theme=dark] .colab-df-quickchart {\n      --bg-color: #3B4455;\n      --fill-color: #D2E3FC;\n      --hover-bg-color: #434B5C;\n      --hover-fill-color: #FFFFFF;\n      --disabled-bg-color: #3B4455;\n      --disabled-fill-color: #666;\n  }\n\n  .colab-df-quickchart {\n    background-color: var(--bg-color);\n    border: none;\n    border-radius: 50%;\n    cursor: pointer;\n    display: none;\n    fill: var(--fill-color);\n    height: 32px;\n    padding: 0;\n    width: 32px;\n  }\n\n  .colab-df-quickchart:hover {\n    background-color: var(--hover-bg-color);\n    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n    fill: var(--button-hover-fill-color);\n  }\n\n  .colab-df-quickchart-complete:disabled,\n  .colab-df-quickchart-complete:disabled:hover {\n    background-color: var(--disabled-bg-color);\n    fill: var(--disabled-fill-color);\n    box-shadow: none;\n  }\n\n  .colab-df-spinner {\n    border: 2px solid var(--fill-color);\n    border-color: transparent;\n    border-bottom-color: var(--fill-color);\n    animation:\n      spin 1s steps(1) infinite;\n  }\n\n  @keyframes spin {\n    0% {\n      border-color: transparent;\n      border-bottom-color: var(--fill-color);\n      border-left-color: var(--fill-color);\n    }\n    20% {\n      border-color: transparent;\n      border-left-color: var(--fill-color);\n      border-top-color: var(--fill-color);\n    }\n    30% {\n      border-color: transparent;\n      border-left-color: var(--fill-color);\n      border-top-color: var(--fill-color);\n      border-right-color: var(--fill-color);\n    }\n    40% {\n      border-color: transparent;\n      border-right-color: var(--fill-color);\n      border-top-color: var(--fill-color);\n    }\n    60% {\n      border-color: transparent;\n      border-right-color: var(--fill-color);\n    }\n    80% {\n      border-color: transparent;\n      border-right-color: var(--fill-color);\n      border-bottom-color: var(--fill-color);\n    }\n    90% {\n      border-color: transparent;\n      border-bottom-color: var(--fill-color);\n    }\n  }\n</style>\n\n  <script>\n    async function quickchart(key) {\n      const quickchartButtonEl =\n        document.querySelector('#' + key + ' button');\n      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n      quickchartButtonEl.classList.add('colab-df-spinner');\n      try {\n        const charts = await google.colab.kernel.invokeFunction(\n            'suggestCharts', [key], {});\n      } catch (error) {\n        console.error('Error during call to suggestCharts:', error);\n      }\n      quickchartButtonEl.classList.remove('colab-df-spinner');\n      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n    }\n    (() => {\n      let quickchartButtonEl =\n        document.querySelector('#df-a5b2c845-4f6e-4022-b145-611c07499cb9 button');\n      quickchartButtonEl.style.display =\n        google.colab.kernel.accessAllowed ? 'block' : 'none';\n    })();\n  </script>\n</div>`;\n        parentElement.appendChild(quickchartButtonContainerElement);\n      }\n\n      appendQuickchartButton(table);\n    "
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As you can see below, the dataset is pretty balanced in terms of the number of examples of each type of instruction (also true for the full dataset with 20,000 rows)."
      ],
      "metadata": {
        "id": "5eIyMPoSGFAN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# num_self_sufficient = (df['input'] == '').sum()\n",
        "num_need_context = main_df.shape[0]\n",
        "# print(num_need_context)\n",
        "# We are only using 100 rows of this dataset for this webinar\n",
        "print(f\"Total number of examples in the dataset: {main_df.shape[0]}\")\n",
        "\n",
        "# print(f\"% of examples that are self-sufficient: {round(num_self_sufficient/main_df.shape[0] * 100, 2)}\")\n",
        "print(f\"% of examples that are need additional context: {round(num_need_context/main_df.shape[0] * 100, 2)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "id": "REqFr-ngx0et",
        "outputId": "74142f7c-1fe1-4ba5-da94-f51ff23af1eb"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total number of examples in the dataset: 500\n",
            "% of examples that are need additional context: 100.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The other aspect worth noting is the average number of characters in each of the three columns `instruction`, `input` and `output` in the dataset. Typically, every 3-4 characters maps to a *token* (the basic building blocks that language models use to understand and analyze text data), and large language models have a limit on the number of tokens they can take as input.\n",
        "\n",
        "The maximum context length for the base LLaMA-2 model is 4096 tokens. Ludwig automatically truncates texts that are too long for the model, but looking at these sequence lengths, we should be able to fine-tune on full length examples without needing any truncation.\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "tcM8DVrIKXWj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculating the length of each cell in each column\n",
        "df['num_characters_question'] = df['question'].apply(lambda x: len(x))\n",
        "# df['num_characters_input'] = df['input'].apply(lambda x: len(x))\n",
        "df['num_characters_answer'] = df['answer'].apply(lambda x: len(x))\n",
        "\n",
        "# Show Distribution\n",
        "df.hist(column=['num_characters_question', 'num_characters_answer'])\n",
        "\n",
        "# Calculating the average\n",
        "average_chars_instruction = df['num_characters_question'].mean()\n",
        "# average_chars_input = df['num_characters_input'].mean()\n",
        "average_chars_output = df['num_characters_answer'].mean()\n",
        "\n",
        "print(f'Average number of tokens in the instruction column: {(average_chars_instruction / 3):.0f}')\n",
        "# print(f'Average number of tokens in the input column: {(average_chars_input / 3):.0f}')\n",
        "print(f'Average number of tokens in the output column: {(average_chars_output / 3):.0f}', end=\"\\n\\n\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 504
        },
        "id": "_3x3stdDKoCT",
        "outputId": "cbf47858-e391-4c52-fecc-a070abd6f9c6"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Average number of tokens in the instruction column: 15\n",
            "Average number of tokens in the output column: 216\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are three different fine-tuning approaches in Ludwig:\n",
        "\n",
        "1. **Full Fine-Tuning**:\n",
        "\n",
        "- Involves training the entire pre-trained model on new data from scratch.\n",
        "- All model layers and parameters are updated during fine-tuning.\n",
        "- Can lead to high accuracy but requires a significant amount of computational resources and time.\n",
        "- Runs the risk of catastrophic forgetting: occasionally, since we are updating all of the weights in the model, this process can lead to the algorithm inadvertently losing knowledge of its past tasks, i.e., the knowledge it gained during pretraining. The outcome may vary, with the algorithm experiencing heightened error margins in some cases, while in others, it might completely erase the memory of a specific task leading to terrible performance.\n",
        "- Best suited when the target task is significantly different from the original pre-training task.\n",
        "\n",
        "2. **Parameter Efficient Fine-Tuning (PEFT), e.g. LoRA**:\n",
        "\n",
        "- Focuses on updating only a subset of the model's parameters.\n",
        "- Often involves freezing certain layers or parts of the model to avoid catastrophic forgetting, or inserting additional layers that are trainable while keeping the original model's weights frozen.\n",
        "- Can result in faster fine-tuning with fewer computational resources, but might sacrifice some accuracy compared to full fine-tuning.\n",
        "- Includes methods like LoRA, AdaLoRA and Adaption Prompt (LLaMA Adapter)\n",
        "- Suitable when the new task shares similarities with the original pre-training task.\n",
        "\n",
        "3. **Quantization-Based Fine-Tuning (QLoRA)**:\n",
        "\n",
        "- Involves reducing the precision of model parameters (e.g., converting 32-bit floating-point values to 8-bit or 4-bit integers). This reduces the amount of CPU and GPU memory required by either 4x if using 8-bit integers, or 8x if using 4-bit integers.\n",
        "- Typically, since we're changing the weights to 8 or 4 bit integers, we will lose some precision/performance.\n",
        "- This can lead to reduced memory usage and faster inference on hardware with reduced precision support.\n",
        "- Particularly useful when deploying models on resource-constrained devices, such as mobile phones or edge devices.\n",
        "\n",
        "\n",
        "**Today, we're going to fine-tune using method 3 since we only have access to a single T4 GPU with 16GiB of GPU VRAM on Colab.** If you have more compute available, give LoRA based fine-tuning or full fine-tuning a try! Typically this requires 4 GPUs with 24GiB of GPU VRAM on a single node multi-GPU cluster and fine-tuning Deepspeeed.\n",
        "\n",
        "\n",
        "To do this, the new parameters we're introducing are:\n",
        "\n",
        "- `adapter`: The PEFT method we want to use\n",
        "- `quantization`: Load the weights in int4 or int8 to reduce memory overhead.\n",
        "- `trainer`: We enable the `finetune` trainer and can configure a variety of training parameters such as epochs and learning rate.\n",
        "\n",
        "Note, there are a few additional preprocessing parameters we should set to ensure that training runs smoothly:\n",
        "\n",
        "```yaml\n",
        "preprocessing:\n",
        "  global_max_sequence_length: 512\n",
        "  split:\n",
        "    type: random\n",
        "    probabilities:\n",
        "    - 1\n",
        "    - 0\n",
        "    - 0\n",
        "```\n",
        "\n",
        "- Some of the examples in the dataset have long sequences, so we set a `global_max_sequence_length` of 512 to ensure that we do not OOM.\n",
        "- Splits are set up such that we use all of the data for training as evaluation phases are synchronous and will take additional time. In a full training run, we recommend using setting aside some data for the test split for evaluation metrics."
      ],
      "metadata": {
        "id": "oU8iKtqwCc1b"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#!pip install numba\n",
        "#\n",
        "#from numba import cuda\n",
        "#device = cuda.get_current_device()\n",
        "#device.reset()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "id": "EtpQTCvx3nal",
        "outputId": "0695d07d-8df1-46ca-ca75-d18af62bab65"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# using Llama-1, just one epoch\n",
        "\n",
        "if False:\n",
        "  model = None\n",
        "  clear_cache()\n",
        "\n",
        "  qlora_fine_tuning_config = yaml.safe_load(\n",
        "\"\"\"\n",
        "model_type: llm\n",
        "base_model: huggyllama/llama-7b\n",
        "\n",
        "input_features:\n",
        "  - name: question\n",
        "    type: text\n",
        "\n",
        "output_features:\n",
        "  - name: answer\n",
        "    type: text\n",
        "\n",
        "prompt:\n",
        "  template: >-\n",
        "    Below is an instruction that describes a task, paired with an input\n",
        "    that provides further context. Write a response that appropriately\n",
        "    completes the request.\n",
        "\n",
        "    ### question: {question}\n",
        "\n",
        "\n",
        "    ### answer:\n",
        "\n",
        "generation:\n",
        "  temperature: 0.1\n",
        "  max_new_tokens: 512\n",
        "\n",
        "adapter:\n",
        "  type: lora\n",
        "\n",
        "quantization:\n",
        "  bits: 4\n",
        "\n",
        "preprocessing:\n",
        "  global_max_sequence_length: 512\n",
        "  split:\n",
        "    type: random\n",
        "    probabilities:\n",
        "    - 1\n",
        "    - 0\n",
        "    - 0\n",
        "\n",
        "trainer:\n",
        "  type: finetune\n",
        "  epochs: 1\n",
        "  batch_size: 1\n",
        "  eval_batch_size: 2\n",
        "  gradient_accumulation_steps: 16\n",
        "  learning_rate: 0.0004\n",
        "  learning_rate_scheduler:\n",
        "    warmup_fraction: 0.03\n",
        "\"\"\"\n",
        ")\n",
        "\n",
        "  model = LudwigModel(config=qlora_fine_tuning_config, logging_level=logging.INFO)\n",
        "  results = model.train(dataset=main_df)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "640b6c0f-7c48-401e-9120-b2d37ebb3740",
        "id": "VaBd80sg3n1d"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Using Llama-2, 5 epochs\n",
        "\n",
        "model2 = None\n",
        "clear_cache()\n",
        "\n",
        "qlora_fine_tuning_config2 = yaml.safe_load(\n",
        "\"\"\"\n",
        "model_type: llm\n",
        "base_model: meta-llama/Llama-2-7b-hf\n",
        "\n",
        "input_features:\n",
        "  - name: question\n",
        "    type: text\n",
        "\n",
        "output_features:\n",
        "  - name: answer\n",
        "    type: text\n",
        "\n",
        "prompt:\n",
        "  template: >-\n",
        "    Below is an instruction that describes a task, paired with an input\n",
        "    that provides further context. Write a response that appropriately\n",
        "    completes the request.\n",
        "\n",
        "    ### question: {question}\n",
        "\n",
        "\n",
        "    ### answer:\n",
        "\n",
        "generation:\n",
        "  temperature: 0.1\n",
        "  max_new_tokens: 512\n",
        "\n",
        "adapter:\n",
        "  type: lora\n",
        "\n",
        "quantization:\n",
        "  bits: 4\n",
        "\n",
        "preprocessing:\n",
        "  global_max_sequence_length: 512\n",
        "  split:\n",
        "    type: random\n",
        "    probabilities:\n",
        "    - 1\n",
        "    - 0\n",
        "    - 0\n",
        "\n",
        "trainer:\n",
        "  type: finetune\n",
        "  epochs: 5\n",
        "  batch_size: 1\n",
        "  eval_batch_size: 2\n",
        "  gradient_accumulation_steps: 16\n",
        "  learning_rate: 0.0004\n",
        "  learning_rate_scheduler:\n",
        "    warmup_fraction: 0.03\n",
        "\"\"\"\n",
        ")\n",
        "\n",
        "model2 = LudwigModel(config=qlora_fine_tuning_config2, logging_level=logging.INFO)\n",
        "results2 = model2.train(dataset=main_df)"
      ],
      "metadata": {
        "id": "k-dtCIj73498",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "595a4b9b3e11459b939041adf43e4f5b",
            "d05c3fd567d24d1e82bfc81a5d390761",
            "42c0706b4b70423ba54bac9ef18e49e2",
            "2c1edac5c0dd4acb96bf24c0ae53d157",
            "133263df5e8348c6ab1146c67cd8750d",
            "adee6ed3b99e47ee8891748798d61dfd",
            "b16ba452c3a94fd6a9eebfa605637c44",
            "38135f188f2944ef97a3d2fc2d7a0f65",
            "c4bf8f58d64544b5a237b26c26d2853d",
            "52831548f39744aa8ef6388d6bafead5",
            "1fbea419c6b640698de5625d1afa89ac",
            "51f7cb00a1bd437386d09109d390c3c0",
            "576dcb8e5d514455a577d24380137869",
            "68b38196f0ee4562b281f692ff733701",
            "748a76f4ddba45bc99c2d1510875a892",
            "4e8f9325fed94f9898fe20348cd9f90a",
            "c636953cbc9e4f5abc7df0e2ae360d64",
            "fe5c12a50dd6472d848225f4ddaf1093",
            "ddd182af26ac418fbb1f473c9cf61b27",
            "9afe6667694d4a23abe0d0754f8d3662",
            "e1c2c41b88ef4198827adb3511dde821",
            "d7ea9730a93e47c3a16d2c775b086495",
            "6c4dbf8ac9cf48acb7c544ace6216901",
            "9ba09e0c332149babfd471b32195f260",
            "ab3e60aa968f467096aaa7a6541f329a",
            "4bb1758031a84760a829959fbe66fd66",
            "5c61d3aaf6b34808bdcefd511b460448",
            "f994af8718cd4afcbac161cc7a2bc2c2",
            "f81ab7826c2f4312943b2b04ae3652a1",
            "e794e79ed35c4e82a81d6366c9b15dbf",
            "146ba63c78d84a83b5dfc008ecc3f7cc",
            "1785d0af5fe842b890549c697a9d96e3",
            "622ca9d2d57d4fc7a7f4ab0cb2c9d16d",
            "7923a48bac9c49479cdab7f1a9d3caa3",
            "3d4d6cd6c8854e20b582e48cb2952f07",
            "fcb2c53b52d9459289f7cbb270fcd0a2",
            "5fd064b4e8254165bbbd3c32ecdead75",
            "a1a54ff6d5284a5e9da499b2b0289a28",
            "29a3396d6926445eae50587a8d813794",
            "f64c980723dd479c88058293efe781d0",
            "e8ddcf1677ac43d8868933ace87eaaa0",
            "cf92a6322d37402d9afcd46eda10f6e4",
            "3cb0685e40bb4de29ff7fb6e2d9b9d16",
            "ba470fbdc9b44ddfaf52f51ffd11c651",
            "91b7da766f844564ad2c54c6da09def9",
            "a3e0192207e04b209a664cdd86f4dfd7",
            "407d79bfe1cd4782aab16e0e5fdefee8",
            "178ffd10380940f1a943f79ea9de7f92",
            "2a60b4f0ebac4829b0273798ab40e940",
            "b7165540b438429aab942db315e68cc9",
            "56b63656744c433fa861082e03e32dda",
            "b3c4b1f3578346799154aed528166f6c",
            "290b2567bab34bbaa4372fadc8945861",
            "a772a77f604940c48db48f1c49a44ec0",
            "9bf2f33b5b774827b38117eaace3f890",
            "4875edb91c9643ad9dcef7d6cf85fcaa",
            "d2885421bec64972b7c95599eebd211c",
            "173474c1a07646d690ff8bf3f0727cc3",
            "94cd0be1871a4b09918344dd61d7e0ff",
            "8a2dfaa1ea0c4492ae7b7dd39a64d661",
            "fd0524ed03e341d4b5b92a6096be9698",
            "05a2dcc1f1ee4e2c9698d0bee576c5f7",
            "73b2d743d3444b8ea746bbd5b8d23d5a",
            "55c4441adc5744c99d57cb571563f082",
            "9f172f8146e945d18d4b544b0e5d99a2",
            "885c4f614b124d46a87707432479388c"
          ]
        },
        "outputId": "e7c97157-b930-45ff-f6cc-f52b6004edad"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.utils.print_utils:╒════════════════════════╕\n",
            "INFO:ludwig.utils.print_utils:│ EXPERIMENT DESCRIPTION │\n",
            "INFO:ludwig.utils.print_utils:╘════════════════════════╛\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.api:╒══════════════════╤═════════════════════════════════════════════════════════════════════════════════════════╕\n",
            "│ Experiment name  │ api_experiment                                                                          │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ Model name       │ run                                                                                     │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ Output directory │ /content/results/api_experiment_run_4                                                   │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ ludwig_version   │ '0.8.3'                                                                                 │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ command          │ ('/usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py -f '                 │\n",
            "│                  │  '/root/.local/share/jupyter/runtime/kernel-e8339515-868d-4ae2-ae16-fb2037f567b8.json') │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ random_seed      │ 42                                                                                      │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ data_format      │ \"<class 'pandas.core.frame.DataFrame'>\"                                                 │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ torch_version    │ '2.0.1+cu118'                                                                           │\n",
            "├──────────────────┼─────────────────────────────────────────────────────────────────────────────────────────┤\n",
            "│ compute          │ {'gpu_type': 'Tesla T4', 'gpus_per_node': 1, 'num_nodes': 1}                            │\n",
            "╘══════════════════╧═════════════════════════════════════════════════════════════════════════════════════════╛\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.utils.print_utils:╒═══════════════╕\n",
            "INFO:ludwig.utils.print_utils:│ LUDWIG CONFIG │\n",
            "INFO:ludwig.utils.print_utils:╘═══════════════╛\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.api:User-specified config (with upgrades):\n",
            "\n",
            "INFO:ludwig.api:{   'adapter': {'type': 'lora'},\n",
            "    'base_model': 'meta-llama/Llama-2-7b-hf',\n",
            "    'generation': {'max_new_tokens': 512, 'temperature': 0.1},\n",
            "    'input_features': [{'name': 'question', 'type': 'text'}],\n",
            "    'ludwig_version': '0.8.3',\n",
            "    'model_type': 'llm',\n",
            "    'output_features': [{'name': 'answer', 'type': 'text'}],\n",
            "    'preprocessing': {   'global_max_sequence_length': 512,\n",
            "                         'split': {   'probabilities': [1, 0, 0],\n",
            "                                      'type': 'random'}},\n",
            "    'prompt': {   'template': 'Below is an instruction that describes a task, '\n",
            "                              'paired with an input that provides further '\n",
            "                              'context. Write a response that appropriately '\n",
            "                              'completes the request.\\n'\n",
            "                              '### question: {question}\\n'\n",
            "                              '\\n'\n",
            "                              '### answer:'},\n",
            "    'quantization': {'bits': 4},\n",
            "    'trainer': {   'batch_size': 1,\n",
            "                   'epochs': 5,\n",
            "                   'eval_batch_size': 2,\n",
            "                   'gradient_accumulation_steps': 16,\n",
            "                   'learning_rate': 0.0004,\n",
            "                   'learning_rate_scheduler': {'warmup_fraction': 0.03},\n",
            "                   'type': 'finetune'}}\n",
            "INFO:ludwig.api:\n",
            "Full config saved to:\n",
            "/content/results/api_experiment_run_4/api_experiment/model/model_hyperparameters.json\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.utils.print_utils:╒═══════════════╕\n",
            "INFO:ludwig.utils.print_utils:│ PREPROCESSING │\n",
            "INFO:ludwig.utils.print_utils:╘═══════════════╛\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.data.preprocessing:No cached dataset found at /content/d43f0bc8557811eea9920242ac1c000c.training.hdf5. Preprocessing the dataset.\n",
            "INFO:ludwig.data.preprocessing:Using full dataframe\n",
            "INFO:ludwig.data.preprocessing:Building dataset (it may take a while)\n",
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n",
            "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n",
            "INFO:ludwig.features.text_feature:Max length of feature 'None': 80 (without start and stop symbols)\n",
            "INFO:ludwig.features.text_feature:Setting max length using dataset: 82 (including start and stop symbols)\n",
            "INFO:ludwig.features.text_feature:max sequence length is 82 for feature 'None'\n",
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n",
            "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n",
            "INFO:ludwig.features.text_feature:Max length of feature 'answer': 412 (without start and stop symbols)\n",
            "INFO:ludwig.features.text_feature:Setting max length using dataset: 414 (including start and stop symbols)\n",
            "INFO:ludwig.features.text_feature:max sequence length is 414 for feature 'answer'\n",
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n",
            "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n",
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n",
            "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n",
            "INFO:ludwig.data.preprocessing:Building dataset: DONE\n",
            "INFO:ludwig.data.cache.manager:Writing preprocessed training set cache to /content/d43f0bc8557811eea9920242ac1c000c.training.hdf5\n",
            "INFO:ludwig.data.cache.manager:Writing preprocessed validation set cache to /content/d43f0bc8557811eea9920242ac1c000c.validation.hdf5\n",
            "INFO:ludwig.data.cache.manager:Writing preprocessed test set cache to /content/d43f0bc8557811eea9920242ac1c000c.test.hdf5\n",
            "INFO:ludwig.data.cache.manager:Writing train set metadata to /content/d43f0bc8557811eea9920242ac1c000c.meta.json\n",
            "WARNING:ludwig.data.preprocessing:Validation set empty. If this is unintentional, please check the preprocessing configuration.\n",
            "WARNING:ludwig.data.preprocessing:Test set empty. If this is unintentional, please check the preprocessing configuration.\n",
            "INFO:ludwig.api:\n",
            "Dataset Statistics\n",
            "INFO:ludwig.api:╒═══════════╤═══════════════╤════════════════════╕\n",
            "│ Dataset   │   Size (Rows) │ Size (In Memory)   │\n",
            "╞═══════════╪═══════════════╪════════════════════╡\n",
            "│ Training  │           500 │ 117.31 Kb          │\n",
            "╘═══════════╧═══════════════╧════════════════════╛\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.utils.print_utils:╒═══════╕\n",
            "INFO:ludwig.utils.print_utils:│ MODEL │\n",
            "INFO:ludwig.utils.print_utils:╘═══════╛\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.api:Warnings and other logs:\n",
            "INFO:ludwig.models.llm:Loading large language model...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)fetensors.index.json:   0%|          | 0.00/26.8k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "595a4b9b3e11459b939041adf43e4f5b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "51f7cb00a1bd437386d09109d390c3c0"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)of-00002.safetensors:   0%|          | 0.00/9.98G [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "6c4dbf8ac9cf48acb7c544ace6216901"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)of-00002.safetensors:   0%|          | 0.00/3.50G [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "7923a48bac9c49479cdab7f1a9d3caa3"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "91b7da766f844564ad2c54c6da09def9"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)neration_config.json:   0%|          | 0.00/188 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "4875edb91c9643ad9dcef7d6cf85fcaa"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.models.llm:Done.\n",
            "/usr/local/lib/python3.10/dist-packages/transformers/generation/configuration_utils.py:362: UserWarning: `do_sample` is set to `False`. However, `temperature` is set to `0.1` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `temperature`. This was detected when initializing the generation config instance, which means the corresponding file may hold incorrect parameterization and should be fixed.\n",
            "  warnings.warn(\n",
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n",
            "INFO:ludwig.models.llm:==================================================\n",
            "INFO:ludwig.models.llm:Trainable Parameter Summary For Fine-Tuning\n",
            "INFO:ludwig.models.llm:Fine-tuning with adapter: lora\n",
            "INFO:ludwig.models.llm:==================================================\n",
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.utils.print_utils:╒══════════╕\n",
            "INFO:ludwig.utils.print_utils:│ TRAINING │\n",
            "INFO:ludwig.utils.print_utils:╘══════════╛\n",
            "INFO:ludwig.utils.print_utils:\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "trainable params: 4,194,304 || all params: 6,742,609,920 || trainable%: 0.06220594176090199\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.trainers.trainer:Creating fresh model training run.\n",
            "INFO:ludwig.trainers.trainer:Training for 2500 step(s), approximately 5 epoch(s).\n",
            "INFO:ludwig.trainers.trainer:Early stopping policy: 5 round(s) of evaluation, or 2500 step(s), approximately 5 epoch(s).\n",
            "\n",
            "INFO:ludwig.trainers.trainer:Starting with step 0, epoch: 0\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  20%|█▉        | 499/2500 [06:25<27:31,  1.21it/s, loss=0.0739]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.data.batcher.random_access:Last batch in epoch only has 1 sample and will be dropped.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  20%|██        | 500/2500 [06:26<37:06,  1.11s/it, loss=0.0896]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.trainers.trainer:\n",
            "Running evaluation for step: 500, epoch: 1\n",
            "INFO:ludwig.trainers.trainer:Evaluation took 0.3510s\n",
            "\n",
            "INFO:ludwig.utils.metrics_printed_table:╒═══════════════════════╤════════════╤══════════════╤════════╕\n",
            "│                       │      train │ validation   │ test   │\n",
            "╞═══════════════════════╪════════════╪══════════════╪════════╡\n",
            "│ bleu                  │     0.1715 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ char_error_rate       │     0.5632 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ loss                  │     1.4330 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ next_token_perplexity │ 18523.3965 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ perplexity            │ 32001.1621 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_fmeasure       │     0.5455 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_precision      │     0.4839 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_recall         │     0.6250 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_fmeasure       │     0.2752 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_precision      │     0.2439 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_recall         │     0.3158 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_fmeasure       │     0.5000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_precision      │     0.4435 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_recall         │     0.5729 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_fmeasure    │     0.5273 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_precision   │     0.4677 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_recall      │     0.6042 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ sequence_accuracy     │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ token_accuracy        │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ word_error_rate       │     0.8261 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ combined_loss         │     1.4330 │              │        │\n",
            "╘═══════════════════════╧════════════╧══════════════╧════════╛\n",
            "INFO:ludwig.trainers.trainer:Saving model.\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  40%|███▉      | 998/2500 [13:06<18:50,  1.33it/s, loss=0.0475]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.data.batcher.random_access:Last batch in epoch only has 1 sample and will be dropped.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  40%|████      | 1000/2500 [13:08<21:54,  1.14it/s, loss=0.06]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.trainers.trainer:\n",
            "Running evaluation for step: 1000, epoch: 2\n",
            "INFO:ludwig.trainers.trainer:Evaluation took 0.1164s\n",
            "\n",
            "INFO:ludwig.utils.metrics_printed_table:╒═══════════════════════╤════════════╤══════════════╤════════╕\n",
            "│                       │      train │ validation   │ test   │\n",
            "╞═══════════════════════╪════════════╪══════════════╪════════╡\n",
            "│ bleu                  │     0.1830 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ char_error_rate       │     0.5935 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ loss                  │     0.9608 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ next_token_perplexity │ 17595.0781 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ perplexity            │ 31998.2930 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_fmeasure       │     0.5658 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_precision      │     0.4674 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_recall         │     0.7167 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_fmeasure       │     0.3200 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_precision      │     0.2637 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_recall         │     0.4068 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_fmeasure       │     0.5395 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_precision      │     0.4457 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_recall         │     0.6833 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_fmeasure    │     0.5526 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_precision   │     0.4565 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_recall      │     0.7000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ sequence_accuracy     │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ token_accuracy        │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ word_error_rate       │     0.9333 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ combined_loss         │     0.9608 │              │        │\n",
            "╘═══════════════════════╧════════════╧══════════════╧════════╛\n",
            "INFO:ludwig.trainers.trainer:Saving model.\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  60%|█████▉    | 1497/2500 [19:47<12:44,  1.31it/s, loss=0.0449]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.data.batcher.random_access:Last batch in epoch only has 1 sample and will be dropped.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  60%|██████    | 1500/2500 [19:50<15:22,  1.08it/s, loss=0.0339]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.trainers.trainer:\n",
            "Running evaluation for step: 1500, epoch: 3\n",
            "INFO:ludwig.trainers.trainer:Evaluation took 0.1217s\n",
            "\n",
            "INFO:ludwig.utils.metrics_printed_table:╒═══════════════════════╤════════════╤══════════════╤════════╕\n",
            "│                       │      train │ validation   │ test   │\n",
            "╞═══════════════════════╪════════════╪══════════════╪════════╡\n",
            "│ bleu                  │     0.4152 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ char_error_rate       │     0.4606 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ loss                  │     0.5430 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ next_token_perplexity │ 15094.0791 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ perplexity            │ 32000.9805 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_fmeasure       │     0.6887 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_precision      │     0.5840 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_recall         │     0.8391 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_fmeasure       │     0.5524 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_precision      │     0.4677 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_recall         │     0.6744 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_fmeasure       │     0.6792 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_precision      │     0.5760 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_recall         │     0.8276 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_fmeasure    │     0.6887 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_precision   │     0.5840 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_recall      │     0.8391 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ sequence_accuracy     │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ token_accuracy        │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ word_error_rate       │     0.6207 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ combined_loss         │     0.5430 │              │        │\n",
            "╘═══════════════════════╧════════════╧══════════════╧════════╛\n",
            "INFO:ludwig.trainers.trainer:Saving model.\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  80%|███████▉  | 1996/2500 [26:29<08:18,  1.01it/s, loss=0.0277]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.data.batcher.random_access:Last batch in epoch only has 1 sample and will be dropped.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training:  80%|████████  | 2000/2500 [26:34<11:17,  1.36s/it, loss=0.0634]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.trainers.trainer:\n",
            "Running evaluation for step: 2000, epoch: 4\n",
            "INFO:ludwig.trainers.trainer:Evaluation took 0.1082s\n",
            "\n",
            "INFO:ludwig.utils.metrics_printed_table:╒═══════════════════════╤════════════╤══════════════╤════════╕\n",
            "│                       │      train │ validation   │ test   │\n",
            "╞═══════════════════════╪════════════╪══════════════╪════════╡\n",
            "│ bleu                  │     0.2083 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ char_error_rate       │     0.4301 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ loss                  │     1.0143 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ next_token_perplexity │ 17007.6230 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ perplexity            │ 31411.6367 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_fmeasure       │     0.5831 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_precision      │     0.5284 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_recall         │     0.6503 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_fmeasure       │     0.3407 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_precision      │     0.3086 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_recall         │     0.3803 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_fmeasure       │     0.5517 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_precision      │     0.5000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_recall         │     0.6154 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_fmeasure    │     0.5705 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_precision   │     0.5170 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_recall      │     0.6364 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ sequence_accuracy     │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ token_accuracy        │     0.0148 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ word_error_rate       │     0.6596 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ combined_loss         │     1.0143 │              │        │\n",
            "╘═══════════════════════╧════════════╧══════════════╧════════╛\n",
            "INFO:ludwig.trainers.trainer:Saving model.\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training: 100%|█████████▉| 2495/2500 [33:15<00:04,  1.10it/s, loss=0.032]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.data.batcher.random_access:Last batch in epoch only has 1 sample and will be dropped.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training: 100%|██████████| 2500/2500 [33:20<00:00,  1.14s/it, loss=0.0311]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.trainers.trainer:\n",
            "Running evaluation for step: 2500, epoch: 5\n",
            "INFO:ludwig.trainers.trainer:Evaluation took 0.1188s\n",
            "\n",
            "INFO:ludwig.utils.metrics_printed_table:╒═══════════════════════╤════════════╤══════════════╤════════╕\n",
            "│                       │      train │ validation   │ test   │\n",
            "╞═══════════════════════╪════════════╪══════════════╪════════╡\n",
            "│ bleu                  │     0.4355 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ char_error_rate       │     0.3980 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ loss                  │     0.4976 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ next_token_perplexity │ 15583.1260 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ perplexity            │ 32001.0078 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_fmeasure       │     0.7167 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_precision      │     0.6370 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge1_recall         │     0.8190 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_fmeasure       │     0.5462 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_precision      │     0.4851 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rouge2_recall         │     0.6250 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_fmeasure       │     0.6833 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_precision      │     0.6074 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeL_recall         │     0.7810 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_fmeasure    │     0.7167 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_precision   │     0.6370 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ rougeLsum_recall      │     0.8190 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ sequence_accuracy     │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ token_accuracy        │     0.0000 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ word_error_rate       │     0.5048 │              │        │\n",
            "├───────────────────────┼────────────┼──────────────┼────────┤\n",
            "│ combined_loss         │     0.4976 │              │        │\n",
            "╘═══════════════════════╧════════════╧══════════════╧════════╛\n",
            "INFO:ludwig.trainers.trainer:Saving model.\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\rTraining: 100%|██████████| 2500/2500 [33:21<00:00,  1.25it/s, loss=0.0311]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.utils.print_utils:\n",
            "INFO:ludwig.utils.print_utils:╒══════════╕\n",
            "INFO:ludwig.utils.print_utils:│ FINISHED │\n",
            "INFO:ludwig.utils.print_utils:╘══════════╛\n",
            "INFO:ludwig.utils.print_utils:\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Perform Inference\n",
        "\n",
        "We can now use the model we fine-tuned above to make predictions on some test examples to see whether fine-tuning the large language model improve its ability to follow instructions/the tasks we're asking it to perform."
      ],
      "metadata": {
        "id": "1YkjcqMFMkwT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "test_examples = pd.DataFrame([\n",
        "      {\n",
        "          \"question\":\"What structure is classified as a definite lie algebra?\",\n",
        "          },\n",
        "\n",
        "      {\n",
        "          \"question\":\"What type of laser is used to study infrared?\",\n",
        "          },\n",
        "      {\n",
        "          \"question\":\"What type of detector detects photon arrival?\",\n",
        "          },\n",
        "      {\n",
        "          \"question\":\"Can a qw be made shapeless?\",\n",
        "          },\n",
        "      {\n",
        "          \"question\":\"Which of the following is the only finite width of quark gluon plasma bags?\",\n",
        "          },\n",
        "      {\n",
        "          \"question\":\"Which phase is associated with a hexagon of hexagons diffraction?\",\n",
        "          },\n",
        "      {\n",
        "          \"question\":\"Where is the companion galaxy?\",\n",
        "          },\n",
        "\n",
        "])\n",
        "\n",
        "#predictions = model.predict(test_examples)[0]\n",
        "predictions = model2.predict(test_examples)[0]\n",
        "\n",
        "#print( predictions )\n",
        "#print( len(test_examples['question']), len(predictions['answer_response']) )\n",
        "\n",
        "for input_with_prediction in zip(test_examples['question'], predictions['answer_response']):\n",
        "  print(f\"Instruction: {input_with_prediction[0]}\")\n",
        "  print(f\"Generated Output: {input_with_prediction[1][0]}\")\n",
        "  print(\"\\n\\n\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ql8BdQm7Mora",
        "outputId": "6ccef967-f71e-4ab1-e24c-0cb0018a74a3"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "  <style>\n",
              "    pre {\n",
              "        white-space: pre-wrap;\n",
              "    }\n",
              "  </style>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n",
            "Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\rPrediction:   0%|          | 0/1 [00:00<?, ?it/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.models.llm:For generating text, using: GenerationConfig {\n",
            "  \"max_length\": 32,\n",
            "  \"max_new_tokens\": 512,\n",
            "  \"temperature\": 0.1,\n",
            "  \"transformers_version\": \"4.33.2\"\n",
            "}\n",
            "\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: what structure is classified as a definite lie algebra?\n",
            "\n",
            "### answer:\n",
            "/usr/local/lib/python3.10/dist-packages/transformers/generation/configuration_utils.py:362: UserWarning: `do_sample` is set to `False`. However, `temperature` is set to `0.1` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `temperature`.\n",
            "  warnings.warn(\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: what structure is classified as a definite lie algebra?\n",
            "\n",
            "### answer: a definite lie algebra is a lie algebra over a field that has a fixed basis. this means that the lie algebra has a set of generators that satisfy a set of defining relations, and these relations are independent of the chosen basis. the structure of a definite lie algebra is fully determined by the structure of its basis and the defining relations.\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: what type of laser is used to study infrared?\n",
            "\n",
            "### answer:\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: what type of laser is used to study infrared?\n",
            "\n",
            "### answer: the most commonly used lasers for infrared studies are near-infrared lasers, which operate at wavelengths between 700 nm and 1000 nm. these lasers are used in a variety of infrared applications, including spectroscopy, imaging, and remote sensing.\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: what type of detector detects photon arrival?\n",
            "\n",
            "### answer:\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: what type of detector detects photon arrival?\n",
            "\n",
            "### answer: a photon detector that is commonly used to detect the arrival of photons is a photomultiplier. it converts the energy of a single photon into an electrical signal that can be measured and analyzed. other types of detectors, such as scintillators or semiconductor detectors, can also be used for photon detection in certain applications.\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: can a qw be made shapeless?\n",
            "\n",
            "### answer:\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: can a qw be made shapeless?\n",
            "\n",
            "### answer: no, a qw (quantum walk) cannot be made shapeless because it is a mathematical concept that describes the behavior of particles in a quantum system. it does not have a shape or form that can be changed.\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: which of the following is the only finite width of quark gluon plasma bags?\n",
            "\n",
            "### answer:\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: which of the following is the only finite width of quark gluon plasma bags?\n",
            "\n",
            "### answer: the only finite width of quark-gluon plasma bags is called the \"horizon\", which is typically on the order of a few hundred fermi (about 1 fm) in size. this is due to the fact that the quark-gluon plasma is believed to be a nearly perfect fluid, which means that it is not possible for it to have any kind of localized structure or boundaries.\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: which phase is associated with a hexagon of hexagons diffraction?\n",
            "\n",
            "### answer:\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: which phase is associated with a hexagon of hexagons diffraction?\n",
            "\n",
            "### answer: the phase associated with a hexagon of hexagons in a diffraction pattern is called the \"br(110)\" phase. this phase is typically observed in a diffraction pattern with a scattering angle of 110°.\n",
            "INFO:ludwig.models.llm:Decoded text inputs for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: where is the companion galaxy?\n",
            "\n",
            "### answer:\n",
            "INFO:ludwig.models.llm:Decoded generated output for the first example in batch: below is an instruction that describes a task, paired with an input that provides further context. write a response that appropriately completes the request.\n",
            "### question: where is the companion galaxy?\n",
            "\n",
            "### answer: the companion galaxy would depend on the specific object you are referring to. in general, companion galaxies are often found orbiting or merging with one another. they can be located in any part of the universe, depending on the object they are orbiting or merging with.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Prediction: 100%|██████████| 1/1 [00:54<00:00, 54.58s/it]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:ludwig.utils.tokenizers:Loaded HuggingFace implementation of meta-llama/Llama-2-7b-hf tokenizer\n",
            "WARNING:ludwig.utils.tokenizers:No padding token id found. Using eos_token as pad_token.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Instruction: What structure is classified as a definite lie algebra?\n",
            "Generated Output: a definite lie algebra is a lie algebra over a field that has a fixed basis. this means that the lie algebra has a set of generators that satisfy a set of defining relations, and these relations are independent of the chosen basis. the structure of a definite lie algebra is fully determined by the structure of its basis and the defining relations.\n",
            "\n",
            "\n",
            "\n",
            "Instruction: What type of laser is used to study infrared?\n",
            "Generated Output: the most commonly used lasers for infrared studies are near-infrared lasers, which operate at wavelengths between 700 nm and 1000 nm. these lasers are used in a variety of infrared applications, including spectroscopy, imaging, and remote sensing.\n",
            "\n",
            "\n",
            "\n",
            "Instruction: What type of detector detects photon arrival?\n",
            "Generated Output: a photon detector that is commonly used to detect the arrival of photons is a photomultiplier. it converts the energy of a single photon into an electrical signal that can be measured and analyzed. other types of detectors, such as scintillators or semiconductor detectors, can also be used for photon detection in certain applications.\n",
            "\n",
            "\n",
            "\n",
            "Instruction: Can a qw be made shapeless?\n",
            "Generated Output: no, a qw (quantum walk) cannot be made shapeless because it is a mathematical concept that describes the behavior of particles in a quantum system. it does not have a shape or form that can be changed.\n",
            "\n",
            "\n",
            "\n",
            "Instruction: Which of the following is the only finite width of quark gluon plasma bags?\n",
            "Generated Output: the only finite width of quark-gluon plasma bags is called the \"horizon\", which is typically on the order of a few hundred fermi (about 1 fm) in size. this is due to the fact that the quark-gluon plasma is believed to be a nearly perfect fluid, which means that it is not possible for it to have any kind of localized structure or boundaries.\n",
            "\n",
            "\n",
            "\n",
            "Instruction: Which phase is associated with a hexagon of hexagons diffraction?\n",
            "Generated Output: the phase associated with a hexagon of hexagons in a diffraction pattern is called the \"br(110)\" phase. this phase is typically observed in a diffraction pattern with a scattering angle of 110°.\n",
            "\n",
            "\n",
            "\n",
            "Instruction: Where is the companion galaxy?\n",
            "Generated Output: the companion galaxy would depend on the specific object you are referring to. in general, companion galaxies are often found orbiting or merging with one another. they can be located in any part of the universe, depending on the object they are orbiting or merging with.\n",
            "\n",
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/ludwig/features/feature_utils.py:102: RuntimeWarning: divide by zero encountered in log\n",
            "  return np.sum(np.log(sequence_probabilities))\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Observations From QLoRA Fine-Tuning** 🔎\n",
        "- Even when we just fine-tune the model on 100 examples from our dataset (which only takes about 4 minutes), it significantly improves the model on our task 🔥\n",
        "- The answers are not perfect when we just use 100 examples, but if we inspect the *logic* in the response, we can see that it is 95% of the way there. This is SIGNIFICANTLY better than before - there is no repetition and the actual code aspects of the answers are all correct.\n",
        "- The partial errors such as `sierp` instead of `arrray` etc indicate that we need to train on a larger amount of data for the model to better learn how to follow instructions and not make these kinds of mistakes."
      ],
      "metadata": {
        "id": "alN-aOHQBLhH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "If you're looking for a managed solution to handle all of the hassle of figuring out the right compute for your fine-tuning task, ensuring that they always succeed without CPU or GPU out-of-memory errors, and be able to rapidly deploy them for fast real-time inference, check out [Predibase](https://www.predibase.com/)."
      ],
      "metadata": {
        "id": "gVOUrJTBD2Io"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# upload the model\n",
        "# !ludwig upload hf_hub --repo_id mflechl/math_physics_50k --model_path /content/results/api_experiment_run"
      ],
      "metadata": {
        "id": "J5YTm1UgMp27"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load saved model\n",
        "\n",
        "from peft import PeftModel, PeftConfig\n",
        "from transformers import AutoModelForCausalLM\n",
        "\n",
        "config = PeftConfig.from_pretrained(\"results/api_experiment_run/model/model_weights/\")\n",
        "model = AutoModelForCausalLM.from_pretrained(\"meta-llama/Llama-2-7b-hf\")\n",
        "model = PeftModel.from_pretrained(model, \"results/api_experiment_run/model/model_weights/\")"
      ],
      "metadata": {
        "id": "lgGfBBBjPT_7"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}