diff --git "a/Conv5d_Experiments.ipynb" "b/Conv5d_Experiments.ipynb" new file mode 100644--- /dev/null +++ "b/Conv5d_Experiments.ipynb" @@ -0,0 +1,84079 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "provenance": [], + "machine_shape": "hm", + "gpuType": "G4" + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "language_info": { + "name": "python" + }, + "accelerator": "GPU", + "widgets": { + "application/vnd.jupyter.widget-state+json": { + "af7741af1af84bd7a81fbb7d3df7a41a": { + "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_f890ba6cefd04309be792af187a14778", + "IPY_MODEL_ed379350acdf445aa85df8ffb3f7a1f1", + "IPY_MODEL_ab6bb3be570143c89474b6be3082fb5b" + ], + "layout": "IPY_MODEL_cfe32c9e840c42bbbef91c80b7b8b886" + } + }, + "f890ba6cefd04309be792af187a14778": { + "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_e9a4b0f6b6eb44e5aee20c32efeea4ab", + "placeholder": "​", + "style": "IPY_MODEL_e932c785787047a98f376680c2928142", + "value": "README.md: " + } + }, + "ed379350acdf445aa85df8ffb3f7a1f1": { + "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_247fe9ef5fc64d25bde33a8116724852", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_34ffabfc9e164f6293bac0bbfac25b14", + "value": 1 + } + }, + "ab6bb3be570143c89474b6be3082fb5b": { + "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_e126fdbf7d2b4b1398cf92e696170a61", + "placeholder": "​", + "style": "IPY_MODEL_d65e49eaa33e454486f1a7f9624d6d06", + "value": " 5.23k/? [00:00<00:00, 1.40MB/s]" + } + }, + "cfe32c9e840c42bbbef91c80b7b8b886": { + "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 + } + }, + "e9a4b0f6b6eb44e5aee20c32efeea4ab": { + "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 + } + }, + "e932c785787047a98f376680c2928142": { + "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": "" + } + }, + "247fe9ef5fc64d25bde33a8116724852": { + "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": "20px" + } + }, + "34ffabfc9e164f6293bac0bbfac25b14": { + "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": "" + } + }, + "e126fdbf7d2b4b1398cf92e696170a61": { + "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 + } + }, + "d65e49eaa33e454486f1a7f9624d6d06": { + "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": "" + } + }, + "6b4725bd99ee4693b47742b4b95033e5": { + "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_784f4114300447a292918ae6cc8274dc", + "IPY_MODEL_d641312e61fe4238bdfd40819b512383", + "IPY_MODEL_19ee6a55eace4f76aa561fe0078b3e18" + ], + "layout": "IPY_MODEL_b7152bd38a6d4388b99acd735364be70" + } + }, + "784f4114300447a292918ae6cc8274dc": { + "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_4c24f92ca35b411f9a2b554523c526c0", + "placeholder": "​", + "style": "IPY_MODEL_91dd7aab738a4c928f93508641381d5b", + "value": "config.json: 100%" + } + }, + "d641312e61fe4238bdfd40819b512383": { + "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_088f7a64855744d3a195f5057e8053b2", + "max": 570, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_72f5b45060a845eaa5776557e0ddcf47", + "value": 570 + } + }, + "19ee6a55eace4f76aa561fe0078b3e18": { + "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_8e9b2ef85269406a91b7b6e593fce474", + "placeholder": "​", + "style": "IPY_MODEL_507c6f4a0089432da303ef03dbbe93ea", + "value": " 570/570 [00:00<00:00, 182kB/s]" + } + }, + "b7152bd38a6d4388b99acd735364be70": { + "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 + } + }, + "4c24f92ca35b411f9a2b554523c526c0": { + "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 + } + }, + "91dd7aab738a4c928f93508641381d5b": { + "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": "" + } + }, + "088f7a64855744d3a195f5057e8053b2": { + "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 + } + }, + "72f5b45060a845eaa5776557e0ddcf47": { + "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": "" + } + }, + "8e9b2ef85269406a91b7b6e593fce474": { + "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 + } + }, + "507c6f4a0089432da303ef03dbbe93ea": { + "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": "" + } + }, + "b708de183007485a84fce02f0f918398": { + "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_ce78e2bc90ad423d9180680f724d7107", + "IPY_MODEL_43d9a583867d4d45a6a3291ec6bb3581", + "IPY_MODEL_e4131698cac54bfba1100e8d8328aa97" + ], + "layout": "IPY_MODEL_3ae0042c5c89410295860b1ff13c7db3" + } + }, + "ce78e2bc90ad423d9180680f724d7107": { + "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_cf806b5ba0aa40a78be556958c52d9bf", + "placeholder": "​", + "style": "IPY_MODEL_dad22612bf4e4958bc0e7af71d35308e", + "value": "model.safetensors: 100%" + } + }, + "43d9a583867d4d45a6a3291ec6bb3581": { + "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_3121edee7f724f82bc3921482672f841", + "max": 440449768, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ce763fde05bb456cb6c9220f10757607", + "value": 440449768 + } + }, + "e4131698cac54bfba1100e8d8328aa97": { + "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_591022de240148fc88f5f02c007ce6ea", + "placeholder": "​", + "style": "IPY_MODEL_2f4c623aecf44199b3cddef3fcc4140e", + "value": " 440M/440M [00:01<00:00, 475MB/s]" + } + }, + "3ae0042c5c89410295860b1ff13c7db3": { + "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 + } + }, + "cf806b5ba0aa40a78be556958c52d9bf": { + "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 + } + }, + "dad22612bf4e4958bc0e7af71d35308e": { + "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": "" + } + }, + "3121edee7f724f82bc3921482672f841": { + "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 + } + }, + "ce763fde05bb456cb6c9220f10757607": { + "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": "" + } + }, + "591022de240148fc88f5f02c007ce6ea": { + "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 + } + }, + "2f4c623aecf44199b3cddef3fcc4140e": { + "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": "" + } + }, + "216c15222c874c359070555738926819": { + "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_3158000d5861479097563eea8c76029e", + "IPY_MODEL_686b256c884a4ecdba4115c417192e14", + "IPY_MODEL_876b5a1218a64365ad500bd53e1850fe" + ], + "layout": "IPY_MODEL_d15678c48b684b81b7daf2f98f161d26" + } + }, + "3158000d5861479097563eea8c76029e": { + "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_1dba758a1c164477be0e3835c498accd", + "placeholder": "​", + "style": "IPY_MODEL_32286444007e41118322a8170daba070", + "value": "Loading weights: 100%" + } + }, + "686b256c884a4ecdba4115c417192e14": { + "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_177b1c0d87bb424797a8c371c7624ec8", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_70797a5f20a54b9099adf3aa23549fd1", + "value": 199 + } + }, + "876b5a1218a64365ad500bd53e1850fe": { + "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_a4c9cbf026004f59bb547584584ed47d", + "placeholder": "​", + "style": "IPY_MODEL_c28367a57ae84981acb2b4173c28cb45", + "value": " 199/199 [00:00<00:00, 4290.46it/s, Materializing param=pooler.dense.weight]" + } + }, + "d15678c48b684b81b7daf2f98f161d26": { + "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 + } + }, + "1dba758a1c164477be0e3835c498accd": { + "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 + } + }, + "32286444007e41118322a8170daba070": { + "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": "" + } + }, + "177b1c0d87bb424797a8c371c7624ec8": { + "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 + } + }, + "70797a5f20a54b9099adf3aa23549fd1": { + "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": "" + } + }, + "a4c9cbf026004f59bb547584584ed47d": { + "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 + } + }, + "c28367a57ae84981acb2b4173c28cb45": { + "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": "" + } + }, + "9989d1226f6a420a9ba12cfb39eb6bd3": { + "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_a669a3fd19fa47bfaf03ed56515c68ec", + "IPY_MODEL_058e48a251d348378acbd5a51df74e86", + "IPY_MODEL_ef1c325ae9d545efb78c569d1ded84c6" + ], + "layout": "IPY_MODEL_6025ac339fd34cd38fc444cdf44299f3" + } + }, + "a669a3fd19fa47bfaf03ed56515c68ec": { + "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_49475ec1c57e4da3a4c149de7a68b787", + "placeholder": "​", + "style": "IPY_MODEL_a8001a46a8e34afca224c48e13a751e8", + "value": "tokenizer_config.json: 100%" + } + }, + "058e48a251d348378acbd5a51df74e86": { + "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_7ed9424968cd48c69762e2c2796293fa", + "max": 48, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_964f870da0dd41d9879cda99e2583c64", + "value": 48 + } + }, + "ef1c325ae9d545efb78c569d1ded84c6": { + "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_c2e096c6cc174454a9c7ea5b0c611ed9", + "placeholder": "​", + "style": "IPY_MODEL_ccac7943ff9b42eba4fefc79cf8e93aa", + "value": " 48.0/48.0 [00:00<00:00, 16.0kB/s]" + } + }, + "6025ac339fd34cd38fc444cdf44299f3": { + "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 + } + }, + "49475ec1c57e4da3a4c149de7a68b787": { + "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 + } + }, + "a8001a46a8e34afca224c48e13a751e8": { + "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": "" + } + }, + "7ed9424968cd48c69762e2c2796293fa": { + "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 + } + }, + "964f870da0dd41d9879cda99e2583c64": { + "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": "" + } + }, + "c2e096c6cc174454a9c7ea5b0c611ed9": { + "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 + } + }, + "ccac7943ff9b42eba4fefc79cf8e93aa": { + "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": "" + } + }, + "249e4535faaa4866a374f6fea46b57a2": { + "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_0ef63ff1dc114ebeb4e5f3388e396faa", + "IPY_MODEL_57d2d9dcbf7c497b9b07312688914a75", + "IPY_MODEL_0c8bb082dd924f43809a544fbf6e22ef" + ], + "layout": "IPY_MODEL_b77cb0b275d8476bae59fc6c17d47115" + } + }, + "0ef63ff1dc114ebeb4e5f3388e396faa": { + "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_dc088f102c904774ac170cd157a18cf0", + "placeholder": "​", + "style": "IPY_MODEL_03599554e1214fa59c53e3a65075ff55", + "value": "vocab.txt: " + } + }, + "57d2d9dcbf7c497b9b07312688914a75": { + "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_0e84ed64f2f44ed69213374cc614d494", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_cbc3ef4fc6e1461db49506d61eee4f21", + "value": 1 + } + }, + "0c8bb082dd924f43809a544fbf6e22ef": { + "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_55e78b09dde14e1182e27ff963b3f472", + "placeholder": "​", + "style": "IPY_MODEL_767d6e11ebc743b68d23e62960686e61", + "value": " 232k/? [00:00<00:00, 17.0MB/s]" + } + }, + "b77cb0b275d8476bae59fc6c17d47115": { + "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 + } + }, + "dc088f102c904774ac170cd157a18cf0": { + "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 + } + }, + "03599554e1214fa59c53e3a65075ff55": { + "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": "" + } + }, + "0e84ed64f2f44ed69213374cc614d494": { + "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": "20px" + } + }, + "cbc3ef4fc6e1461db49506d61eee4f21": { + "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": "" + } + }, + "55e78b09dde14e1182e27ff963b3f472": { + "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 + } + }, + "767d6e11ebc743b68d23e62960686e61": { + "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": "" + } + }, + "5a91836f467e46959f1b80db2ca6afea": { + "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_b840c5497d154057b7eebe02f47ccc51", + "IPY_MODEL_1672c7add2d544428851a21ed996dd0a", + "IPY_MODEL_4dc0c89c1a724f159f1ca1473dac9f5f" + ], + "layout": "IPY_MODEL_098c25ce400b4293980b2bdbcc9f2bf9" + } + }, + "b840c5497d154057b7eebe02f47ccc51": { + "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_23f8dc075ee94518927e405921fb5d4b", + "placeholder": "​", + "style": "IPY_MODEL_098c1f83a8f24c41ab4cd7e229a56754", + "value": "tokenizer.json: " + } + }, + "1672c7add2d544428851a21ed996dd0a": { + "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_6a3947c2b7c744bfbed33e605327e1e7", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c984888d8311469ebf2c66d83c98c749", + "value": 1 + } + }, + "4dc0c89c1a724f159f1ca1473dac9f5f": { + "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_c665765cb32942448634fb62540aff20", + "placeholder": "​", + "style": "IPY_MODEL_913f6311e9bf4058a2d301295e01aae7", + "value": " 466k/? [00:00<00:00, 43.8MB/s]" + } + }, + "098c25ce400b4293980b2bdbcc9f2bf9": { + "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 + } + }, + "23f8dc075ee94518927e405921fb5d4b": { + "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 + } + }, + "098c1f83a8f24c41ab4cd7e229a56754": { + "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": "" + } + }, + "6a3947c2b7c744bfbed33e605327e1e7": { + "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": "20px" + } + }, + "c984888d8311469ebf2c66d83c98c749": { + "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": "" + } + }, + "c665765cb32942448634fb62540aff20": { + "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 + } + }, + "913f6311e9bf4058a2d301295e01aae7": { + "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": "" + } + }, + "eed503f995d74c449acfb217ce0acd7d": { + "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_584923fc781040288e04c2df23ba6207", + "IPY_MODEL_21209e6cc4af496c96a918ee64f7e114", + "IPY_MODEL_7c82222cb6e54855935571450a66d72c" + ], + "layout": "IPY_MODEL_2b9bf9f8d67f4f31ba5f592f68431da8" + } + }, + "584923fc781040288e04c2df23ba6207": { + "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_06a18d9b194f4d5783ba45f2b674cded", + "placeholder": "​", + "style": "IPY_MODEL_7f237b344f734beb9dbe005c33a553dd", + "value": "config.json: " + } + }, + "21209e6cc4af496c96a918ee64f7e114": { + "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_52cdc94fe2b84df88e54a7c34f5fd38f", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_dfa6dce60ea44b3d8b3a264b90684fc0", + "value": 1 + } + }, + "7c82222cb6e54855935571450a66d72c": { + "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_261b5f47bc1748b98cca6125a96fc8e0", + "placeholder": "​", + "style": "IPY_MODEL_c8efdf5df992463e9cca11c28186d33d", + "value": " 1.19k/? [00:00<00:00, 371kB/s]" + } + }, + "2b9bf9f8d67f4f31ba5f592f68431da8": { + "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 + } + }, + "06a18d9b194f4d5783ba45f2b674cded": { + "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 + } + }, + "7f237b344f734beb9dbe005c33a553dd": { + "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": "" + } + }, + "52cdc94fe2b84df88e54a7c34f5fd38f": { + "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": "20px" + } + }, + "dfa6dce60ea44b3d8b3a264b90684fc0": { + "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": "" + } + }, + "261b5f47bc1748b98cca6125a96fc8e0": { + "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 + } + }, + "c8efdf5df992463e9cca11c28186d33d": { + "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": "" + } + }, + "2c869b1b80574876bb99de5b3535fd2c": { + "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_ba8969ad55e44c638e115599042a7b37", + "IPY_MODEL_8a4fbfc103be43c28d430a40de7d2213", + "IPY_MODEL_e1ae0da775fb4af1bbcb3387c6c3deac" + ], + "layout": "IPY_MODEL_e945c3c437ca4ac69901538823b69203" + } + }, + "ba8969ad55e44c638e115599042a7b37": { + "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_698bca7a9f6c4b0c81e047e49a9aa046", + "placeholder": "​", + "style": "IPY_MODEL_45a1b40fa4a0460abc61e6d82ce14a4c", + "value": "model.safetensors: 100%" + } + }, + "8a4fbfc103be43c28d430a40de7d2213": { + "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_5c6ac9d91f4245c6b96ad0a5d9102e41", + "max": 598635032, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_bb9870743c2c49699d04fc1802b72577", + "value": 598635032 + } + }, + "e1ae0da775fb4af1bbcb3387c6c3deac": { + "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_78b2e7158eb94aedba001fbd14fe98c3", + "placeholder": "​", + "style": "IPY_MODEL_b220cdda21a94f2dbaa3b627fbeb78ee", + "value": " 599M/599M [00:02<00:00, 451MB/s]" + } + }, + "e945c3c437ca4ac69901538823b69203": { + "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 + } + }, + "698bca7a9f6c4b0c81e047e49a9aa046": { + "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 + } + }, + "45a1b40fa4a0460abc61e6d82ce14a4c": { + "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": "" + } + }, + "5c6ac9d91f4245c6b96ad0a5d9102e41": { + "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 + } + }, + "bb9870743c2c49699d04fc1802b72577": { + "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": "" + } + }, + "78b2e7158eb94aedba001fbd14fe98c3": { + "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 + } + }, + "b220cdda21a94f2dbaa3b627fbeb78ee": { + "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": "" + } + }, + "af2a792cfc4a4886afb14f23a69e8e32": { + "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_017d834170154d6caa6fe3fbcad87cc8", + "IPY_MODEL_f94b989f3400401fb93db36bb4ae1e43", + "IPY_MODEL_ef1197d0a7d24adcafb7f62814f2fa64" + ], + "layout": "IPY_MODEL_4e22cb7b7eb8437d94a5f3cd84c00cad" + } + }, + "017d834170154d6caa6fe3fbcad87cc8": { + "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_82fdb2e5fc3540c2adcd4717520b4f05", + "placeholder": "​", + "style": "IPY_MODEL_2e55bc310fd94b5b88a11ed8844cf989", + "value": "Loading weights: 100%" + } + }, + "f94b989f3400401fb93db36bb4ae1e43": { + "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_e57da4bd40da4427a224607b5a17fcdf", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_8c40cdfb5ae544dea8f1458da2d35cb1", + "value": 134 + } + }, + "ef1197d0a7d24adcafb7f62814f2fa64": { + "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_158bc2ca4b304e6e94fcd839a30d47f4", + "placeholder": "​", + "style": "IPY_MODEL_f5e3d0180b7f418693ca40d042674d97", + "value": " 134/134 [00:00<00:00, 4092.66it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "4e22cb7b7eb8437d94a5f3cd84c00cad": { + "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 + } + }, + "82fdb2e5fc3540c2adcd4717520b4f05": { + "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 + } + }, + "2e55bc310fd94b5b88a11ed8844cf989": { + "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": "" + } + }, + "e57da4bd40da4427a224607b5a17fcdf": { + "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 + } + }, + "8c40cdfb5ae544dea8f1458da2d35cb1": { + "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": "" + } + }, + "158bc2ca4b304e6e94fcd839a30d47f4": { + "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 + } + }, + "f5e3d0180b7f418693ca40d042674d97": { + "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": "" + } + }, + "a085e226619c41f6853fef4568929453": { + "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_4fc62e7e1e19420aa7974190a48b76ac", + "IPY_MODEL_f419ed984fe14816bf27f313ae93b7c7", + "IPY_MODEL_2cfb05e56c964bbfae9b4eab1935c4e7" + ], + "layout": "IPY_MODEL_ca7dc6d7370347699c4fdf7a210aa898" + } + }, + "4fc62e7e1e19420aa7974190a48b76ac": { + "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_51cc7cb998b24fa5a62722a363f18fd2", + "placeholder": "​", + "style": "IPY_MODEL_62f7292135d24aeeb9eb4ca2f8fb93e9", + "value": "tokenizer_config.json: " + } + }, + "f419ed984fe14816bf27f313ae93b7c7": { + "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_cc3251b8ba30447abf38e3d1804e4c45", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0e28de4e717e4d26a003b1da684a9b50", + "value": 1 + } + }, + "2cfb05e56c964bbfae9b4eab1935c4e7": { + "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_15878b16affc454d8edf48f9ade837bf", + "placeholder": "​", + "style": "IPY_MODEL_d1952bd013b742b988caf768547b648f", + "value": " 20.8k/? [00:00<00:00, 6.31MB/s]" + } + }, + "ca7dc6d7370347699c4fdf7a210aa898": { + "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 + } + }, + "51cc7cb998b24fa5a62722a363f18fd2": { + "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 + } + }, + "62f7292135d24aeeb9eb4ca2f8fb93e9": { + "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": "" + } + }, + "cc3251b8ba30447abf38e3d1804e4c45": { + "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": "20px" + } + }, + "0e28de4e717e4d26a003b1da684a9b50": { + "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": "" + } + }, + "15878b16affc454d8edf48f9ade837bf": { + "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 + } + }, + "d1952bd013b742b988caf768547b648f": { + "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": "" + } + }, + "d8491c9e16c44cb686845d39f8ac392d": { + "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_772fbd960f124a39b87531b2661ba5b8", + "IPY_MODEL_4b73d09efe854ed99bf962c1f6791d4b", + "IPY_MODEL_425902f67d044118a72cefcb35a4a7da" + ], + "layout": "IPY_MODEL_39804dbc2f224ff49fe4dbaba666bdab" + } + }, + "772fbd960f124a39b87531b2661ba5b8": { + "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_cce9161655c943538a5c485a6c5aeff4", + "placeholder": "​", + "style": "IPY_MODEL_044ce8c81d2a4d67beae9f945068e5cb", + "value": "tokenizer.json: " + } + }, + "4b73d09efe854ed99bf962c1f6791d4b": { + "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_4ca65781df6242a6ab2b1f72a1de9f4a", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_8158cfa78ddb4cf6a6693cb026347a6a", + "value": 1 + } + }, + "425902f67d044118a72cefcb35a4a7da": { + "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_fbdf4d1087614c03af3037b03208df82", + "placeholder": "​", + "style": "IPY_MODEL_c4dbefe5d8e6410db7b2228d897766f6", + "value": " 2.13M/? [00:00<00:00, 40.6MB/s]" + } + }, + "39804dbc2f224ff49fe4dbaba666bdab": { + "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 + } + }, + "cce9161655c943538a5c485a6c5aeff4": { + "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 + } + }, + "044ce8c81d2a4d67beae9f945068e5cb": { + "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": "" + } + }, + "4ca65781df6242a6ab2b1f72a1de9f4a": { + "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": "20px" + } + }, + "8158cfa78ddb4cf6a6693cb026347a6a": { + "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": "" + } + }, + "fbdf4d1087614c03af3037b03208df82": { + "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 + } + }, + "c4dbefe5d8e6410db7b2228d897766f6": { + "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": "" + } + }, + "91a81ac85cd04855b64b721830cb1ac2": { + "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_6ceef1cd00314bdaa54537cb4ccd8634", + "IPY_MODEL_6a345a7d08b54545a02a4bd706705d9a", + "IPY_MODEL_02e8c91e46af4e6bbab16b9a80b675a8" + ], + "layout": "IPY_MODEL_632c821a403d4cbdaa18e9c96848bdfc" + } + }, + "6ceef1cd00314bdaa54537cb4ccd8634": { + "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_108db55efdf049efb551e9397c3d0072", + "placeholder": "​", + "style": "IPY_MODEL_c7d4f9bea5474958ad5975451c65fdd4", + "value": "special_tokens_map.json: 100%" + } + }, + "6a345a7d08b54545a02a4bd706705d9a": { + "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_8db8d51ee82f408793ff80a62c7654eb", + "max": 694, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_20f3700205444e24ae3562944a5dadad", + "value": 694 + } + }, + "02e8c91e46af4e6bbab16b9a80b675a8": { + "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_dec81f3e632045ef9a2b91ec1febb795", + "placeholder": "​", + "style": "IPY_MODEL_a23face7e21b44918c9f1edea5053c08", + "value": " 694/694 [00:00<00:00, 216kB/s]" + } + }, + "632c821a403d4cbdaa18e9c96848bdfc": { + "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 + } + }, + "108db55efdf049efb551e9397c3d0072": { + "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 + } + }, + "c7d4f9bea5474958ad5975451c65fdd4": { + "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": "" + } + }, + "8db8d51ee82f408793ff80a62c7654eb": { + "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 + } + }, + "20f3700205444e24ae3562944a5dadad": { + "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": "" + } + }, + "dec81f3e632045ef9a2b91ec1febb795": { + "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 + } + }, + "a23face7e21b44918c9f1edea5053c08": { + "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": "" + } + }, + "e2a70dcb8e624d7ca612ba7bb8c32807": { + "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_ea3eb28389e64a86af508ad6f2d1a6ac", + "IPY_MODEL_23f1cccb23564eba9e1e53d82ccf3c78", + "IPY_MODEL_36002b606f774b2eb2a9e158acf0f197" + ], + "layout": "IPY_MODEL_d32f2d541d34419e9e2464c2ac664341" + } + }, + "ea3eb28389e64a86af508ad6f2d1a6ac": { + "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_061f46eb38264ecebdad782966a10c47", + "placeholder": "​", + "style": "IPY_MODEL_bebfb41784e1440cb605c43325394e63", + "value": "Loading weights: 100%" + } + }, + "23f1cccb23564eba9e1e53d82ccf3c78": { + "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_e97b733720ea47a3a68ec22e43f9b3e6", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_56454f346b4047c994e068d2de5347b1", + "value": 199 + } + }, + "36002b606f774b2eb2a9e158acf0f197": { + "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_52d8c3faf2484db2881201ae857be33b", + "placeholder": "​", + "style": "IPY_MODEL_2edd21ec2d664f039ba2beb7f91d90d5", + "value": " 199/199 [00:00<00:00, 4250.18it/s, Materializing param=pooler.dense.weight]" + } + }, + "d32f2d541d34419e9e2464c2ac664341": { + "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 + } + }, + "061f46eb38264ecebdad782966a10c47": { + "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 + } + }, + "bebfb41784e1440cb605c43325394e63": { + "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": "" + } + }, + "e97b733720ea47a3a68ec22e43f9b3e6": { + "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 + } + }, + "56454f346b4047c994e068d2de5347b1": { + "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": "" + } + }, + "52d8c3faf2484db2881201ae857be33b": { + "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 + } + }, + "2edd21ec2d664f039ba2beb7f91d90d5": { + "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": "" + } + }, + "83d590542cd046c1973e3134f758a50f": { + "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_3a7256139434453eb33e1feaef040346", + "IPY_MODEL_aef3afc5e6ab47969a396fab50f3f9cb", + "IPY_MODEL_a1e2a7e6f3b24ec5af32a65096f7e77c" + ], + "layout": "IPY_MODEL_8c70676303854eb78eb23f3327774a0f" + } + }, + "3a7256139434453eb33e1feaef040346": { + "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_231321a4c42c4cf1b6eacc1b3fdcd45b", + "placeholder": "​", + "style": "IPY_MODEL_f1183aa32f974ddd97dcb1943b2575c7", + "value": "Loading weights: 100%" + } + }, + "aef3afc5e6ab47969a396fab50f3f9cb": { + "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_3b6cad5748534d88a198a58224d352a1", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_fd3edc5d5df542639132bdb5bf15dd37", + "value": 134 + } + }, + "a1e2a7e6f3b24ec5af32a65096f7e77c": { + "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_181cadcf3bb6486c9710946daf92f28a", + "placeholder": "​", + "style": "IPY_MODEL_0f4c4ff3caed46c1a1ba1dc3fda5357c", + "value": " 134/134 [00:00<00:00, 3974.13it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "8c70676303854eb78eb23f3327774a0f": { + "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 + } + }, + "231321a4c42c4cf1b6eacc1b3fdcd45b": { + "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 + } + }, + "f1183aa32f974ddd97dcb1943b2575c7": { + "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": "" + } + }, + "3b6cad5748534d88a198a58224d352a1": { + "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 + } + }, + "fd3edc5d5df542639132bdb5bf15dd37": { + "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": "" + } + }, + "181cadcf3bb6486c9710946daf92f28a": { + "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 + } + }, + "0f4c4ff3caed46c1a1ba1dc3fda5357c": { + "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": "" + } + }, + "62b084d4045f495f956b8c77b53d979b": { + "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_b95b061016fc466c92c039fde8d782ad", + "IPY_MODEL_0dc3db85b0fb444da5175f161fb4cb64", + "IPY_MODEL_d01aea89ae924779b6e649e7b75c2188" + ], + "layout": "IPY_MODEL_0f73b0865aed4866a5b138fa2891033f" + } + }, + "b95b061016fc466c92c039fde8d782ad": { + "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_634a0196709a4268b2b2dfc035ef2633", + "placeholder": "​", + "style": "IPY_MODEL_2e7bb615b7ad44b286ea675a17c6de1d", + "value": "config.json: 100%" + } + }, + "0dc3db85b0fb444da5175f161fb4cb64": { + "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_5472f2e989744382ae0347b545ee25d6", + "max": 481, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_fbba8d61e96f49c9aec2cde5ab51a8f9", + "value": 481 + } + }, + "d01aea89ae924779b6e649e7b75c2188": { + "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_86940e9440ec464eb5d92496b6a214e5", + "placeholder": "​", + "style": "IPY_MODEL_7f22f17245134bf1b1c2518becce877e", + "value": " 481/481 [00:00<00:00, 169kB/s]" + } + }, + "0f73b0865aed4866a5b138fa2891033f": { + "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 + } + }, + "634a0196709a4268b2b2dfc035ef2633": { + "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 + } + }, + "2e7bb615b7ad44b286ea675a17c6de1d": { + "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": "" + } + }, + "5472f2e989744382ae0347b545ee25d6": { + "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 + } + }, + "fbba8d61e96f49c9aec2cde5ab51a8f9": { + "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": "" + } + }, + "86940e9440ec464eb5d92496b6a214e5": { + "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 + } + }, + "7f22f17245134bf1b1c2518becce877e": { + "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": "" + } + }, + "f2948dca39ba4ecc9d98b1102133ed34": { + "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_c100049742444cf9b85e175b6d2d7430", + "IPY_MODEL_9b3127ed91eb4cfc8785e580c50f4c23", + "IPY_MODEL_4660729f84194c52a80cbf6f29892d56" + ], + "layout": "IPY_MODEL_4c20a430b6da43a1be4080bb8461490e" + } + }, + "c100049742444cf9b85e175b6d2d7430": { + "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_f5200448b77f4964a9956b75988ab47a", + "placeholder": "​", + "style": "IPY_MODEL_61b0266940074f74bcad883c522af1c0", + "value": "model.safetensors: 100%" + } + }, + "9b3127ed91eb4cfc8785e580c50f4c23": { + "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_4633d4dcc2b945eeb75c14a418d4954f", + "max": 498818054, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_41d0cc3eab034873b41711d6464143b4", + "value": 498818054 + } + }, + "4660729f84194c52a80cbf6f29892d56": { + "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_ed42298f6a33403bbc9f6877908ef4b8", + "placeholder": "​", + "style": "IPY_MODEL_26cb536be18c49669d547ef543abd23d", + "value": " 499M/499M [00:01<00:00, 538MB/s]" + } + }, + "4c20a430b6da43a1be4080bb8461490e": { + "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 + } + }, + "f5200448b77f4964a9956b75988ab47a": { + "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 + } + }, + "61b0266940074f74bcad883c522af1c0": { + "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": "" + } + }, + "4633d4dcc2b945eeb75c14a418d4954f": { + "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 + } + }, + "41d0cc3eab034873b41711d6464143b4": { + "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": "" + } + }, + "ed42298f6a33403bbc9f6877908ef4b8": { + "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 + } + }, + "26cb536be18c49669d547ef543abd23d": { + "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": "" + } + }, + "cb5fb6c625ba49e9ab6a03a709725662": { + "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_f9385bf5b575401985f0d05ba3348ef7", + "IPY_MODEL_9392f75ede4a44f7bf880e2e0210783e", + "IPY_MODEL_d594a78f634e4d76940590350ec15e43" + ], + "layout": "IPY_MODEL_70e97d70ab3d4acba3326bb2e47151b4" + } + }, + "f9385bf5b575401985f0d05ba3348ef7": { + "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_5cf478922c4447f5acd97aaf233c0727", + "placeholder": "​", + "style": "IPY_MODEL_ff3c65ba985b4e1f8acf3f95061c54f6", + "value": "Loading weights: 100%" + } + }, + "9392f75ede4a44f7bf880e2e0210783e": { + "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_28ed8de161b2450798b3a1fa04d7e82b", + "max": 197, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0cd1f7c0b786403d8ba40a8008e50517", + "value": 197 + } + }, + "d594a78f634e4d76940590350ec15e43": { + "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_d94e7581fcab4474bd751eb3a8897f3c", + "placeholder": "​", + "style": "IPY_MODEL_2df3c52aec3440c08e56dd9ccbdf2138", + "value": " 197/197 [00:00<00:00, 4189.94it/s, Materializing param=encoder.layer.11.output.dense.weight]" + } + }, + "70e97d70ab3d4acba3326bb2e47151b4": { + "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 + } + }, + "5cf478922c4447f5acd97aaf233c0727": { + "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 + } + }, + "ff3c65ba985b4e1f8acf3f95061c54f6": { + "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": "" + } + }, + "28ed8de161b2450798b3a1fa04d7e82b": { + "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 + } + }, + "0cd1f7c0b786403d8ba40a8008e50517": { + "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": "" + } + }, + "d94e7581fcab4474bd751eb3a8897f3c": { + "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 + } + }, + "2df3c52aec3440c08e56dd9ccbdf2138": { + "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": "" + } + }, + "80433f6dfd8d49949a7070b6cf65edb1": { + "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_9101e148e3db460d89e4d644f612fb48", + "IPY_MODEL_79261c5003774f7e8aa78536da2396e6", + "IPY_MODEL_b6455e90ba3e4723aefc133e2fd4fd1e" + ], + "layout": "IPY_MODEL_17dcf7cbb6904d44a98eef1a232c046d" + } + }, + "9101e148e3db460d89e4d644f612fb48": { + "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_8f683fa10a7643d0bc2acd3bcd11c6e4", + "placeholder": "​", + "style": "IPY_MODEL_e37af08ec462491e8460cec81f75f645", + "value": "tokenizer_config.json: 100%" + } + }, + "79261c5003774f7e8aa78536da2396e6": { + "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_31e200e2295a432a9e2f34df9b1c5d8f", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d8c1777a148740dc9178bd2ce79484b3", + "value": 25 + } + }, + "b6455e90ba3e4723aefc133e2fd4fd1e": { + "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_db4cafd36e9340bead8b71d66558731c", + "placeholder": "​", + "style": "IPY_MODEL_a6f32e2206664bfc9bc10b45003bfc7c", + "value": " 25.0/25.0 [00:00<00:00, 8.40kB/s]" + } + }, + "17dcf7cbb6904d44a98eef1a232c046d": { + "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 + } + }, + "8f683fa10a7643d0bc2acd3bcd11c6e4": { + "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 + } + }, + "e37af08ec462491e8460cec81f75f645": { + "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": "" + } + }, + "31e200e2295a432a9e2f34df9b1c5d8f": { + "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 + } + }, + "d8c1777a148740dc9178bd2ce79484b3": { + "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": "" + } + }, + "db4cafd36e9340bead8b71d66558731c": { + "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 + } + }, + "a6f32e2206664bfc9bc10b45003bfc7c": { + "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": "" + } + }, + "f3a9156ccae24f7982bc2331d9f4efbd": { + "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_889099da7a914f989d176034fc542749", + "IPY_MODEL_618ce6d77e1c49f59294492982f9b8f7", + "IPY_MODEL_4fc77c8e89a44eb59f69dd4507f83500" + ], + "layout": "IPY_MODEL_258e203e3ed3448f89edbb13876d586c" + } + }, + "889099da7a914f989d176034fc542749": { + "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_de0171a8b863438d866eb33cb10bd5a2", + "placeholder": "​", + "style": "IPY_MODEL_0453764ff6594970859447f2b68cfdef", + "value": "vocab.json: " + } + }, + "618ce6d77e1c49f59294492982f9b8f7": { + "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_64e72ab3ec174c2f905f65858ceba074", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_bc59bc91a14c4172aca65ef536d9cb05", + "value": 1 + } + }, + "4fc77c8e89a44eb59f69dd4507f83500": { + "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_611fc78d12e2420e932c5a45d96744a2", + "placeholder": "​", + "style": "IPY_MODEL_cd79f5e46fd94a45a7d7f4c0aac8959b", + "value": " 899k/? [00:00<00:00, 7.66MB/s]" + } + }, + "258e203e3ed3448f89edbb13876d586c": { + "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 + } + }, + "de0171a8b863438d866eb33cb10bd5a2": { + "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 + } + }, + "0453764ff6594970859447f2b68cfdef": { + "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": "" + } + }, + "64e72ab3ec174c2f905f65858ceba074": { + "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": "20px" + } + }, + "bc59bc91a14c4172aca65ef536d9cb05": { + "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": "" + } + }, + "611fc78d12e2420e932c5a45d96744a2": { + "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 + } + }, + "cd79f5e46fd94a45a7d7f4c0aac8959b": { + "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": "" + } + }, + "3c245882f54848338fe6b071b536b713": { + "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_8660f44e8efd473e8c1c695d5b860d49", + "IPY_MODEL_557a696649b04a3295766bc64c092692", + "IPY_MODEL_187b10e26325482c96824ba7417d3a6a" + ], + "layout": "IPY_MODEL_a69ae42a7ecb4db8b409acbe40ba7c0a" + } + }, + "8660f44e8efd473e8c1c695d5b860d49": { + "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_cae4a557f8ca444a98fd91a956212b5f", + "placeholder": "​", + "style": "IPY_MODEL_d4c14d6d688a45e1b0c9a45f5eea0006", + "value": "merges.txt: " + } + }, + "557a696649b04a3295766bc64c092692": { + "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_55c9a9d615fe43cbae080a3aa74522cb", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_3b8261809aaf422caa65a8a1aba6f726", + "value": 1 + } + }, + "187b10e26325482c96824ba7417d3a6a": { + "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_d1cea2635e02449ba64d7533c7c0e772", + "placeholder": "​", + "style": "IPY_MODEL_76dbe1b8b6734ee9b663b64a9756a85f", + "value": " 456k/? [00:00<00:00, 8.88MB/s]" + } + }, + "a69ae42a7ecb4db8b409acbe40ba7c0a": { + "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 + } + }, + "cae4a557f8ca444a98fd91a956212b5f": { + "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 + } + }, + "d4c14d6d688a45e1b0c9a45f5eea0006": { + "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": "" + } + }, + "55c9a9d615fe43cbae080a3aa74522cb": { + "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": "20px" + } + }, + "3b8261809aaf422caa65a8a1aba6f726": { + "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": "" + } + }, + "d1cea2635e02449ba64d7533c7c0e772": { + "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 + } + }, + "76dbe1b8b6734ee9b663b64a9756a85f": { + "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": "" + } + }, + "d21ad66f0bee4ed7a715cf5be3c3de13": { + "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_28deaa25dee94e8cb665323cadebb3ab", + "IPY_MODEL_c9bf3dc381214fc680a67677d38120ca", + "IPY_MODEL_769e0cec00484eddba520ee2573c20fb" + ], + "layout": "IPY_MODEL_c99b0229a251429a867f67786c345126" + } + }, + "28deaa25dee94e8cb665323cadebb3ab": { + "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_b1ba274cdf454d3e84c4f992b3804cfb", + "placeholder": "​", + "style": "IPY_MODEL_26bae17f338e4da3a11e7e7a256541fa", + "value": "tokenizer.json: " + } + }, + "c9bf3dc381214fc680a67677d38120ca": { + "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_f098d81430dc49eda2dd364cb0ad4166", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_59cee6f235a14eecbb01578cfa73e374", + "value": 1 + } + }, + "769e0cec00484eddba520ee2573c20fb": { + "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_b23fe17ac64049c78fd56fe31da7f47b", + "placeholder": "​", + "style": "IPY_MODEL_5ea02354c8d844bfa0e64383b4a48003", + "value": " 1.36M/? [00:00<00:00, 7.98MB/s]" + } + }, + "c99b0229a251429a867f67786c345126": { + "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 + } + }, + "b1ba274cdf454d3e84c4f992b3804cfb": { + "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 + } + }, + "26bae17f338e4da3a11e7e7a256541fa": { + "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": "" + } + }, + "f098d81430dc49eda2dd364cb0ad4166": { + "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": "20px" + } + }, + "59cee6f235a14eecbb01578cfa73e374": { + "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": "" + } + }, + "b23fe17ac64049c78fd56fe31da7f47b": { + "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 + } + }, + "5ea02354c8d844bfa0e64383b4a48003": { + "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": "" + } + }, + "bf0fc650cd764e4bb91845df43652588": { + "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_a127908ce3f04fe2aeef8e3c34eeff79", + "IPY_MODEL_178af9e08e184caa8a6eba114b5412d8", + "IPY_MODEL_ed84b2dfdb7a410ba25b50e639a78e23" + ], + "layout": "IPY_MODEL_3ffbfc7e2d69438cb7ba83ffed391be0" + } + }, + "a127908ce3f04fe2aeef8e3c34eeff79": { + "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_dad41b0099f44d50875daa47b2d7db6f", + "placeholder": "​", + "style": "IPY_MODEL_bbae2ea3bfc94322a3fc960e9ff58bc2", + "value": "config.json: 100%" + } + }, + "178af9e08e184caa8a6eba114b5412d8": { + "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_70a74bf0df9b45b0990db8cbbcf9d0af", + "max": 684, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_2ce31be0bebc4b44b8404d4e81ce9c3c", + "value": 684 + } + }, + "ed84b2dfdb7a410ba25b50e639a78e23": { + "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_043827c1bed044d9a4fe14a353cccbfc", + "placeholder": "​", + "style": "IPY_MODEL_a97d5f96f2af4b42ae08503db12c4ea4", + "value": " 684/684 [00:00<00:00, 247kB/s]" + } + }, + "3ffbfc7e2d69438cb7ba83ffed391be0": { + "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 + } + }, + "dad41b0099f44d50875daa47b2d7db6f": { + "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 + } + }, + "bbae2ea3bfc94322a3fc960e9ff58bc2": { + "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": "" + } + }, + "70a74bf0df9b45b0990db8cbbcf9d0af": { + "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 + } + }, + "2ce31be0bebc4b44b8404d4e81ce9c3c": { + "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": "" + } + }, + "043827c1bed044d9a4fe14a353cccbfc": { + "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 + } + }, + "a97d5f96f2af4b42ae08503db12c4ea4": { + "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": "" + } + }, + "a7326715179e43a78b1efe16165988d1": { + "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_edb27c6faf0c4636a6e7e713d700cec3", + "IPY_MODEL_ef69b38c0e1b4242b2a7d2d3e9382eaf", + "IPY_MODEL_9b96cb040afa4bdc9b5d85c1a20cd592" + ], + "layout": "IPY_MODEL_86defdd6dd354c068eda620264f2b2bb" + } + }, + "edb27c6faf0c4636a6e7e713d700cec3": { + "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_73ee733a1afb4674bd46ba582d504312", + "placeholder": "​", + "style": "IPY_MODEL_c45f9a287be24e8aaa41a93553621038", + "value": "model.safetensors: 100%" + } + }, + "ef69b38c0e1b4242b2a7d2d3e9382eaf": { + "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_d4beb1b020a442ecb7bbe5f144eb9ec1", + "max": 47372894, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_722f0c93b55a42ebbce9b05515155b9a", + "value": 47372894 + } + }, + "9b96cb040afa4bdc9b5d85c1a20cd592": { + "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_f41ab9f3f1a843a0a9d4922f3c891755", + "placeholder": "​", + "style": "IPY_MODEL_97b07779f8634852b3dd2a9ca33aefce", + "value": " 47.4M/47.4M [00:00<00:00, 236MB/s]" + } + }, + "86defdd6dd354c068eda620264f2b2bb": { + "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 + } + }, + "73ee733a1afb4674bd46ba582d504312": { + "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 + } + }, + "c45f9a287be24e8aaa41a93553621038": { + "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": "" + } + }, + "d4beb1b020a442ecb7bbe5f144eb9ec1": { + "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 + } + }, + "722f0c93b55a42ebbce9b05515155b9a": { + "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": "" + } + }, + "f41ab9f3f1a843a0a9d4922f3c891755": { + "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 + } + }, + "97b07779f8634852b3dd2a9ca33aefce": { + "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": "" + } + }, + "29b4b025d9e749cb8828a088a819a78f": { + "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_67192e1b2c8f4996aee8a7ac87730bfa", + "IPY_MODEL_1a7b34cddb3347b8b37bdc2603646055", + "IPY_MODEL_331a16de953840dca34c3370a2907830" + ], + "layout": "IPY_MODEL_748f5114da474866b1b737b0010f0978" + } + }, + "67192e1b2c8f4996aee8a7ac87730bfa": { + "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_9e2659f5996949918dd84b9f7ed010ac", + "placeholder": "​", + "style": "IPY_MODEL_9e306c886520406ab752fedfa23e7758", + "value": "Loading weights: 100%" + } + }, + "1a7b34cddb3347b8b37bdc2603646055": { + "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_482069bc1b1d4fe684f05c73a208a240", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d055c436b5124b4295a9ae3ff1129462", + "value": 25 + } + }, + "331a16de953840dca34c3370a2907830": { + "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_7d14322b6bb64d8aacd68cfdfd543149", + "placeholder": "​", + "style": "IPY_MODEL_b522e34d74d1459cac9fe8218a1f3287", + "value": " 25/25 [00:00<00:00, 2864.49it/s, Materializing param=pooler.weight]" + } + }, + "748f5114da474866b1b737b0010f0978": { + "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 + } + }, + "9e2659f5996949918dd84b9f7ed010ac": { + "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 + } + }, + "9e306c886520406ab752fedfa23e7758": { + "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": "" + } + }, + "482069bc1b1d4fe684f05c73a208a240": { + "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 + } + }, + "d055c436b5124b4295a9ae3ff1129462": { + "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": "" + } + }, + "7d14322b6bb64d8aacd68cfdfd543149": { + "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 + } + }, + "b522e34d74d1459cac9fe8218a1f3287": { + "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": "" + } + }, + "e6bdea9e40b44959a5c58a9e279af2d0": { + "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_b38a2df501c4486da4b0f1ba9b1819ec", + "IPY_MODEL_0cccd0189a4a4268b297b76448ba7bdb", + "IPY_MODEL_10d4198a08ef4995877e63bb3fa528a8" + ], + "layout": "IPY_MODEL_2b78c4d3b6c74687a960d7dad357a132" + } + }, + "b38a2df501c4486da4b0f1ba9b1819ec": { + "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_561bd0ea372d4f2987c340894fda2f6f", + "placeholder": "​", + "style": "IPY_MODEL_e8dbe71c495d4d2ea289117544889a66", + "value": "tokenizer_config.json: 100%" + } + }, + "0cccd0189a4a4268b297b76448ba7bdb": { + "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_4395d639c16d4c6caaa274f2161bef23", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_dea61b9776f34c099930b4d0b5dd4032", + "value": 25 + } + }, + "10d4198a08ef4995877e63bb3fa528a8": { + "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_9da65bd197794b81a12b6ab59a36577b", + "placeholder": "​", + "style": "IPY_MODEL_c22e76554ca149288e3bd7c731dcc6af", + "value": " 25.0/25.0 [00:00<00:00, 9.77kB/s]" + } + }, + "2b78c4d3b6c74687a960d7dad357a132": { + "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 + } + }, + "561bd0ea372d4f2987c340894fda2f6f": { + "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 + } + }, + "e8dbe71c495d4d2ea289117544889a66": { + "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": "" + } + }, + "4395d639c16d4c6caaa274f2161bef23": { + "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 + } + }, + "dea61b9776f34c099930b4d0b5dd4032": { + "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": "" + } + }, + "9da65bd197794b81a12b6ab59a36577b": { + "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 + } + }, + "c22e76554ca149288e3bd7c731dcc6af": { + "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": "" + } + }, + "b34f02689eb74f10b587d15bd355c4f0": { + "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_ea1d6d9fc1f14dfd9bd766ced0b7408f", + "IPY_MODEL_7b6e7f8037544f9a8bd6987b20e26ce6", + "IPY_MODEL_cc0d6b6e28414bba9455ae019e1cf077" + ], + "layout": "IPY_MODEL_9902f2dc7f8043ca8a54a751a11e16a5" + } + }, + "ea1d6d9fc1f14dfd9bd766ced0b7408f": { + "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_248e37ccd3b04d58be04cfd1f743f00f", + "placeholder": "​", + "style": "IPY_MODEL_d30dc98d3c1543f1ab4576d42c884820", + "value": "spiece.model: 100%" + } + }, + "7b6e7f8037544f9a8bd6987b20e26ce6": { + "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_d58ef9fc339a4c078f59a48dccf14269", + "max": 760289, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0963bfd1678342bfad73f57615dc3253", + "value": 760289 + } + }, + "cc0d6b6e28414bba9455ae019e1cf077": { + "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_a183e56ca7704eecaa2ee15ca5a1c62b", + "placeholder": "​", + "style": "IPY_MODEL_995fb042fcc74a2aae45c52fda9b1f8b", + "value": " 760k/760k [00:00<00:00, 20.8MB/s]" + } + }, + "9902f2dc7f8043ca8a54a751a11e16a5": { + "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 + } + }, + "248e37ccd3b04d58be04cfd1f743f00f": { + "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 + } + }, + "d30dc98d3c1543f1ab4576d42c884820": { + "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": "" + } + }, + "d58ef9fc339a4c078f59a48dccf14269": { + "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 + } + }, + "0963bfd1678342bfad73f57615dc3253": { + "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": "" + } + }, + "a183e56ca7704eecaa2ee15ca5a1c62b": { + "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 + } + }, + "995fb042fcc74a2aae45c52fda9b1f8b": { + "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": "" + } + }, + "e396f3c394284383a8c02fb5509dab40": { + "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_78d7e796eb934b95bf408f2de7ee6183", + "IPY_MODEL_b3edbb4a0ffd42f1a7422fa9b0c598e8", + "IPY_MODEL_1795f93e690748b99caaa90d805b72d2" + ], + "layout": "IPY_MODEL_e97177add3844f72a71a624c71403309" + } + }, + "78d7e796eb934b95bf408f2de7ee6183": { + "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_bced4455899849e9b549ee5242389e0d", + "placeholder": "​", + "style": "IPY_MODEL_1915dfeb013a4cbebd14f0c244bf0d59", + "value": "tokenizer.json: " + } + }, + "b3edbb4a0ffd42f1a7422fa9b0c598e8": { + "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_95cab53c46a240fcbe9966bda5ce26ce", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0f99cb8305ac4246b8b74588f726ac33", + "value": 1 + } + }, + "1795f93e690748b99caaa90d805b72d2": { + "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_2ce464af80ab4c25a4d15a74569c12ad", + "placeholder": "​", + "style": "IPY_MODEL_f3826f5dbd2846baa00931b8c559b5dc", + "value": " 1.31M/? [00:00<00:00, 11.9MB/s]" + } + }, + "e97177add3844f72a71a624c71403309": { + "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 + } + }, + "bced4455899849e9b549ee5242389e0d": { + "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 + } + }, + "1915dfeb013a4cbebd14f0c244bf0d59": { + "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": "" + } + }, + "95cab53c46a240fcbe9966bda5ce26ce": { + "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": "20px" + } + }, + "0f99cb8305ac4246b8b74588f726ac33": { + "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": "" + } + }, + "2ce464af80ab4c25a4d15a74569c12ad": { + "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 + } + }, + "f3826f5dbd2846baa00931b8c559b5dc": { + "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": "" + } + }, + "7fe531947a98458fb61ad5ccece362ad": { + "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_ff0ce2d8a69b4024aee9313567801eac", + "IPY_MODEL_520c5393a85c40f5a0905c3d9fad6c9c", + "IPY_MODEL_e872a1f8826041c7b91e152bd2b70831" + ], + "layout": "IPY_MODEL_5e2a2b6f85bc4688bd5524ce74618239" + } + }, + "ff0ce2d8a69b4024aee9313567801eac": { + "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_ec6422eeba5e43a68d97ca9a480b04f1", + "placeholder": "​", + "style": "IPY_MODEL_85b7d5f09eab4f8898d39d1125908c97", + "value": "config.json: 100%" + } + }, + "520c5393a85c40f5a0905c3d9fad6c9c": { + "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_6b594c0e28f7472ba5d8449eb2a814b3", + "max": 483, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_9225d3431b9e4bf8878670033cd37048", + "value": 483 + } + }, + "e872a1f8826041c7b91e152bd2b70831": { + "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_d50f3aa199ac448e9e5af9f6de771990", + "placeholder": "​", + "style": "IPY_MODEL_38164c5851694c978f4cae3599b46b89", + "value": " 483/483 [00:00<00:00, 170kB/s]" + } + }, + "5e2a2b6f85bc4688bd5524ce74618239": { + "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 + } + }, + "ec6422eeba5e43a68d97ca9a480b04f1": { + "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 + } + }, + "85b7d5f09eab4f8898d39d1125908c97": { + "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": "" + } + }, + "6b594c0e28f7472ba5d8449eb2a814b3": { + "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 + } + }, + "9225d3431b9e4bf8878670033cd37048": { + "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": "" + } + }, + "d50f3aa199ac448e9e5af9f6de771990": { + "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 + } + }, + "38164c5851694c978f4cae3599b46b89": { + "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": "" + } + }, + "a66cbab44e1f44dc96447a4f7b00f798": { + "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_abc873dbcbe64a81bb0fee6608294bdf", + "IPY_MODEL_8f942478ee1e457cb61e07abba8edc70", + "IPY_MODEL_bcb36fb962584748836b7ca776775d2a" + ], + "layout": "IPY_MODEL_f07ab6e1039c49d18d66a217afa29897" + } + }, + "abc873dbcbe64a81bb0fee6608294bdf": { + "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_3241865cbc6c442a95d4d38e82aa25ac", + "placeholder": "​", + "style": "IPY_MODEL_4893c83a88044ca39c5f00b686c09a03", + "value": "model.safetensors: 100%" + } + }, + "8f942478ee1e457cb61e07abba8edc70": { + "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_d34cce7e6ca44bd0a3664df642a1f7dc", + "max": 267954768, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ece74c977a73489f9efdec2fa7e778ba", + "value": 267954768 + } + }, + "bcb36fb962584748836b7ca776775d2a": { + "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_8047d5dbbcad42aa8969cd6aa2eca828", + "placeholder": "​", + "style": "IPY_MODEL_5c2dfffdc8b04d1885d9ce1f02a7d0d1", + "value": " 268M/268M [00:01<00:00, 271MB/s]" + } + }, + "f07ab6e1039c49d18d66a217afa29897": { + "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 + } + }, + "3241865cbc6c442a95d4d38e82aa25ac": { + "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 + } + }, + "4893c83a88044ca39c5f00b686c09a03": { + "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": "" + } + }, + "d34cce7e6ca44bd0a3664df642a1f7dc": { + "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 + } + }, + "ece74c977a73489f9efdec2fa7e778ba": { + "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": "" + } + }, + "8047d5dbbcad42aa8969cd6aa2eca828": { + "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 + } + }, + "5c2dfffdc8b04d1885d9ce1f02a7d0d1": { + "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": "" + } + }, + "c3e0a5e0dc2547fd96e384b4142094fb": { + "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_9e0098e4f054480fae8dccb70eb5fc16", + "IPY_MODEL_dda522f33a23466db417679e65673e24", + "IPY_MODEL_aeafa57dec3442ccb07287952e31bf16" + ], + "layout": "IPY_MODEL_bd51f8731dce43a0a73da9133037a9f7" + } + }, + "9e0098e4f054480fae8dccb70eb5fc16": { + "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_9c98f555b91d4edd9e1e84043e60edc7", + "placeholder": "​", + "style": "IPY_MODEL_0e381b6c796443a0a5734254bbe09ea5", + "value": "Loading weights: 100%" + } + }, + "dda522f33a23466db417679e65673e24": { + "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_95c3a67d20a74da58e619915e17d25ba", + "max": 100, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ee5c30553b3e48588295b90295d7943a", + "value": 100 + } + }, + "aeafa57dec3442ccb07287952e31bf16": { + "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_23f17bc680d34ad194a26b8b93693e1b", + "placeholder": "​", + "style": "IPY_MODEL_d47da607c91640ed93859e5be6137d01", + "value": " 100/100 [00:00<00:00, 3697.02it/s, Materializing param=transformer.layer.5.sa_layer_norm.weight]" + } + }, + "bd51f8731dce43a0a73da9133037a9f7": { + "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 + } + }, + "9c98f555b91d4edd9e1e84043e60edc7": { + "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 + } + }, + "0e381b6c796443a0a5734254bbe09ea5": { + "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": "" + } + }, + "95c3a67d20a74da58e619915e17d25ba": { + "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 + } + }, + "ee5c30553b3e48588295b90295d7943a": { + "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": "" + } + }, + "23f17bc680d34ad194a26b8b93693e1b": { + "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 + } + }, + "d47da607c91640ed93859e5be6137d01": { + "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": "" + } + }, + "77727da895e64e33afb03b21abf632a9": { + "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_f9d1285d6c9f47d4a5586a4002b3461f", + "IPY_MODEL_8b8b9f5e4dc741da8fcdaa667fe4cffa", + "IPY_MODEL_2e7d00e6737d419daa31f38427df40d2" + ], + "layout": "IPY_MODEL_40dc7545a0b547a396d666457057cc34" + } + }, + "f9d1285d6c9f47d4a5586a4002b3461f": { + "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_760719f9e92349ca9360e900e589271e", + "placeholder": "​", + "style": "IPY_MODEL_925232a874d840fb8ae504c5cdcbec2f", + "value": "tokenizer_config.json: 100%" + } + }, + "8b8b9f5e4dc741da8fcdaa667fe4cffa": { + "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_6bce40c76023457281f6838b3ba74f69", + "max": 48, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_54538f8399184581837b414057445b49", + "value": 48 + } + }, + "2e7d00e6737d419daa31f38427df40d2": { + "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_2ab7eee807c746ffa44fdff3fec9cd2f", + "placeholder": "​", + "style": "IPY_MODEL_a81bad4243984c99a1f2fbe0355f72c3", + "value": " 48.0/48.0 [00:00<00:00, 17.0kB/s]" + } + }, + "40dc7545a0b547a396d666457057cc34": { + "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 + } + }, + "760719f9e92349ca9360e900e589271e": { + "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 + } + }, + "925232a874d840fb8ae504c5cdcbec2f": { + "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": "" + } + }, + "6bce40c76023457281f6838b3ba74f69": { + "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 + } + }, + "54538f8399184581837b414057445b49": { + "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": "" + } + }, + "2ab7eee807c746ffa44fdff3fec9cd2f": { + "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 + } + }, + "a81bad4243984c99a1f2fbe0355f72c3": { + "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": "" + } + }, + "f6585ffef775480c867f8ac095bdc292": { + "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_c70e8217531d4778a4d7341a8604e514", + "IPY_MODEL_afe1af71c4dd44929c2a1b2a8d02f6e9", + "IPY_MODEL_aede4d97d7ef4e899ec63d29b8fbfe9c" + ], + "layout": "IPY_MODEL_2a4db79fbbaa4e7b885c0c8dfbf3a9f6" + } + }, + "c70e8217531d4778a4d7341a8604e514": { + "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_00f59dabbd1a4377ba0b0bbebd7523a6", + "placeholder": "​", + "style": "IPY_MODEL_49342ca5408f4ba1985b64e307802b99", + "value": "vocab.txt: " + } + }, + "afe1af71c4dd44929c2a1b2a8d02f6e9": { + "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_68988518c92545c0a2fa1900e6ad30e1", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_2ccded8d8b0a4317947c3f04813bdcc0", + "value": 1 + } + }, + "aede4d97d7ef4e899ec63d29b8fbfe9c": { + "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_345dc059805148d5ae130956fe755da8", + "placeholder": "​", + "style": "IPY_MODEL_b9767ace4e39498188862d285e1c7234", + "value": " 232k/? [00:00<00:00, 6.39MB/s]" + } + }, + "2a4db79fbbaa4e7b885c0c8dfbf3a9f6": { + "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 + } + }, + "00f59dabbd1a4377ba0b0bbebd7523a6": { + "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 + } + }, + "49342ca5408f4ba1985b64e307802b99": { + "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": "" + } + }, + "68988518c92545c0a2fa1900e6ad30e1": { + "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": "20px" + } + }, + "2ccded8d8b0a4317947c3f04813bdcc0": { + "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": "" + } + }, + "345dc059805148d5ae130956fe755da8": { + "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 + } + }, + "b9767ace4e39498188862d285e1c7234": { + "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": "" + } + }, + "346ce9b77d454356b569ec3383f1492f": { + "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_4c3d20d7ede34cb0934bc1ca0b7ec4bd", + "IPY_MODEL_18449ba42d6f4997aecdcf7ccc0253c7", + "IPY_MODEL_e3c98c90f79f4917afbcb712e2408839" + ], + "layout": "IPY_MODEL_831f5133f327465da02e6ac03a77e03e" + } + }, + "4c3d20d7ede34cb0934bc1ca0b7ec4bd": { + "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_9baf758c7b324cba91f3af4d9107d944", + "placeholder": "​", + "style": "IPY_MODEL_b639b1a63b59473daad7deced5c3d60c", + "value": "tokenizer.json: " + } + }, + "18449ba42d6f4997aecdcf7ccc0253c7": { + "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_933a7bd9df2c43a3b71ecd02d897aa9f", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_23593ce9f71d4424b7e089d78111baab", + "value": 1 + } + }, + "e3c98c90f79f4917afbcb712e2408839": { + "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_25b604ffaf9d49878f4326ffcc90fc11", + "placeholder": "​", + "style": "IPY_MODEL_0edd02992610457397183dcfe4969f01", + "value": " 466k/? [00:00<00:00, 5.18MB/s]" + } + }, + "831f5133f327465da02e6ac03a77e03e": { + "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 + } + }, + "9baf758c7b324cba91f3af4d9107d944": { + "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 + } + }, + "b639b1a63b59473daad7deced5c3d60c": { + "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": "" + } + }, + "933a7bd9df2c43a3b71ecd02d897aa9f": { + "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": "20px" + } + }, + "23593ce9f71d4424b7e089d78111baab": { + "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": "" + } + }, + "25b604ffaf9d49878f4326ffcc90fc11": { + "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 + } + }, + "0edd02992610457397183dcfe4969f01": { + "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": "" + } + }, + "0eaf1b9b3faf4564b2ea751d3292a62b": { + "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_61f98e3de9f54cd8bb6ac5abf94c4547", + "IPY_MODEL_e81f9655321c435f898aaf1d2bef667c", + "IPY_MODEL_ff6d3c3b5f6c45a58746baf241ab8f69" + ], + "layout": "IPY_MODEL_6311dfcc5ede419cbebd064cb42af75a" + } + }, + "61f98e3de9f54cd8bb6ac5abf94c4547": { + "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_5f3e8530131045459d831a4cd75d752d", + "placeholder": "​", + "style": "IPY_MODEL_80acec4b777947bf9481d1b6292178c0", + "value": "Loading weights: 100%" + } + }, + "e81f9655321c435f898aaf1d2bef667c": { + "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_1a6e19bfe5244157a110d80615352827", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_96d62c3895ec40449c75d8527dc6f1e5", + "value": 199 + } + }, + "ff6d3c3b5f6c45a58746baf241ab8f69": { + "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_3e71024a339a4938a94205fcbb4ba530", + "placeholder": "​", + "style": "IPY_MODEL_2038b2fbe7d846178ecf6cf083ca55de", + "value": " 199/199 [00:00<00:00, 4295.56it/s, Materializing param=pooler.dense.weight]" + } + }, + "6311dfcc5ede419cbebd064cb42af75a": { + "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 + } + }, + "5f3e8530131045459d831a4cd75d752d": { + "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 + } + }, + "80acec4b777947bf9481d1b6292178c0": { + "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": "" + } + }, + "1a6e19bfe5244157a110d80615352827": { + "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 + } + }, + "96d62c3895ec40449c75d8527dc6f1e5": { + "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": "" + } + }, + "3e71024a339a4938a94205fcbb4ba530": { + "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 + } + }, + "2038b2fbe7d846178ecf6cf083ca55de": { + "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": "" + } + }, + "abdd244173644b379041034854e908fa": { + "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_122a0e0816f945ceb1942cd3d10f51f6", + "IPY_MODEL_6ef64c471dc94f609440fb4c4ed0f9e4", + "IPY_MODEL_449a39840fac45a9bf33b016a4ecde66" + ], + "layout": "IPY_MODEL_af961b31ee0e4cabac70dd4e61cd2f05" + } + }, + "122a0e0816f945ceb1942cd3d10f51f6": { + "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_e372742ba260499b875618da8b0b62df", + "placeholder": "​", + "style": "IPY_MODEL_a3d8b3f6224a4f5e85a1b1a8fc116cc0", + "value": "Loading weights: 100%" + } + }, + "6ef64c471dc94f609440fb4c4ed0f9e4": { + "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_3273f54a6fcc4e82aee577b2b508b493", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_631ea49dbaa3495688f16e0b27c90610", + "value": 134 + } + }, + "449a39840fac45a9bf33b016a4ecde66": { + "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_beed45db486d40dbb70fd5a6713fcd5d", + "placeholder": "​", + "style": "IPY_MODEL_4a2611a199da4fef9dafd0c3934e02fe", + "value": " 134/134 [00:00<00:00, 3957.89it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "af961b31ee0e4cabac70dd4e61cd2f05": { + "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 + } + }, + "e372742ba260499b875618da8b0b62df": { + "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 + } + }, + "a3d8b3f6224a4f5e85a1b1a8fc116cc0": { + "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": "" + } + }, + "3273f54a6fcc4e82aee577b2b508b493": { + "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 + } + }, + "631ea49dbaa3495688f16e0b27c90610": { + "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": "" + } + }, + "beed45db486d40dbb70fd5a6713fcd5d": { + "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 + } + }, + "4a2611a199da4fef9dafd0c3934e02fe": { + "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": "" + } + }, + "defad155b30b4076ae5431e8b299a148": { + "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_cac0d30d1ba848f28e2722a7baddf6fe", + "IPY_MODEL_38b121d084cc40e58727c89299acfef3", + "IPY_MODEL_bbe4c87a57804f3d9332abc541c938ed" + ], + "layout": "IPY_MODEL_7754907597864c7386d8234500b1df3b" + } + }, + "cac0d30d1ba848f28e2722a7baddf6fe": { + "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_34c8e8c9f27f4384b33a50f37f0dd97c", + "placeholder": "​", + "style": "IPY_MODEL_65d3b06a14c74e9f944194dbe682d070", + "value": "Loading weights: 100%" + } + }, + "38b121d084cc40e58727c89299acfef3": { + "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_8ff7e87e23e048ce90ca973a803874e9", + "max": 197, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_07d12ebd42d1426091b26eeb77a17d97", + "value": 197 + } + }, + "bbe4c87a57804f3d9332abc541c938ed": { + "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_d77868ac7580404490ca0f80ec607cc3", + "placeholder": "​", + "style": "IPY_MODEL_d666624772a6477d90dce56414e97ac9", + "value": " 197/197 [00:00<00:00, 4109.16it/s, Materializing param=encoder.layer.11.output.dense.weight]" + } + }, + "7754907597864c7386d8234500b1df3b": { + "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 + } + }, + "34c8e8c9f27f4384b33a50f37f0dd97c": { + "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 + } + }, + "65d3b06a14c74e9f944194dbe682d070": { + "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": "" + } + }, + "8ff7e87e23e048ce90ca973a803874e9": { + "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 + } + }, + "07d12ebd42d1426091b26eeb77a17d97": { + "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": "" + } + }, + "d77868ac7580404490ca0f80ec607cc3": { + "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 + } + }, + "d666624772a6477d90dce56414e97ac9": { + "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": "" + } + }, + "8eb0acb8d0174d89ad3869e17c73468c": { + "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_ba39be331d814cb4ac763e3f9fde1032", + "IPY_MODEL_132de8e2ddd541269a810f44cc6fb971", + "IPY_MODEL_e4ce3e29132a45649817df4f7e8341e2" + ], + "layout": "IPY_MODEL_3bf5704b4d6942909ad163b3b91318af" + } + }, + "ba39be331d814cb4ac763e3f9fde1032": { + "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_87dbb92120d84e418a9ba3770b6ddb73", + "placeholder": "​", + "style": "IPY_MODEL_e4eda51e8e614607901352ff01958332", + "value": "Loading weights: 100%" + } + }, + "132de8e2ddd541269a810f44cc6fb971": { + "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_e0234180e8644ba99c472de64827ce69", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_99482875446a40d488ded2b1522820f9", + "value": 25 + } + }, + "e4ce3e29132a45649817df4f7e8341e2": { + "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_a177a905339d494a9938dee8e6446983", + "placeholder": "​", + "style": "IPY_MODEL_c5ed7bba1e6a4cf18e5bc3bda01bcada", + "value": " 25/25 [00:00<00:00, 2791.74it/s, Materializing param=pooler.weight]" + } + }, + "3bf5704b4d6942909ad163b3b91318af": { + "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 + } + }, + "87dbb92120d84e418a9ba3770b6ddb73": { + "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 + } + }, + "e4eda51e8e614607901352ff01958332": { + "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": "" + } + }, + "e0234180e8644ba99c472de64827ce69": { + "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 + } + }, + "99482875446a40d488ded2b1522820f9": { + "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": "" + } + }, + "a177a905339d494a9938dee8e6446983": { + "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 + } + }, + "c5ed7bba1e6a4cf18e5bc3bda01bcada": { + "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": "" + } + }, + "380a95ce1d17446fb7a8370f95e2d62d": { + "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_17ba56553c1545dc8899a7ff580e3b4a", + "IPY_MODEL_43e22d4a71a74918be6c8c4a6dccd928", + "IPY_MODEL_96ddc7a8e7f4482aa863aae9194bc2fa" + ], + "layout": "IPY_MODEL_698be3985bc14d93bdb4f2d8ad7114f9" + } + }, + "17ba56553c1545dc8899a7ff580e3b4a": { + "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_07154f164e574d4ea174183bab680de6", + "placeholder": "​", + "style": "IPY_MODEL_b943e03d5d094af2abf688177557d1b3", + "value": "Loading weights: 100%" + } + }, + "43e22d4a71a74918be6c8c4a6dccd928": { + "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_5daca214c00c40a08aaa0ccb44406d93", + "max": 100, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d473e9f0063341a4871d5c507d012fe7", + "value": 100 + } + }, + "96ddc7a8e7f4482aa863aae9194bc2fa": { + "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_4a39b4a3ddb5468ebeb0f595732d681f", + "placeholder": "​", + "style": "IPY_MODEL_92a8707d70614dada558e2cc8c90b1d7", + "value": " 100/100 [00:00<00:00, 3593.72it/s, Materializing param=transformer.layer.5.sa_layer_norm.weight]" + } + }, + "698be3985bc14d93bdb4f2d8ad7114f9": { + "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 + } + }, + "07154f164e574d4ea174183bab680de6": { + "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 + } + }, + "b943e03d5d094af2abf688177557d1b3": { + "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": "" + } + }, + "5daca214c00c40a08aaa0ccb44406d93": { + "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 + } + }, + "d473e9f0063341a4871d5c507d012fe7": { + "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": "" + } + }, + "4a39b4a3ddb5468ebeb0f595732d681f": { + "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 + } + }, + "92a8707d70614dada558e2cc8c90b1d7": { + "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": "" + } + }, + "445251c2934740f2bf149cd2e7848887": { + "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_94548dbad7fd4c8386087683768b5644", + "IPY_MODEL_8654b6b67b294522a6c6005820745747", + "IPY_MODEL_2da39e52fb8d463bae47fdb295fd95c2" + ], + "layout": "IPY_MODEL_fab849e9e4d34fdcb47fb6f5b37ce8b7" + } + }, + "94548dbad7fd4c8386087683768b5644": { + "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_e71d4fe04ba14e41b3d34012299b33f0", + "placeholder": "​", + "style": "IPY_MODEL_24ea33a9dd494980902f2d2d2d3c2f2e", + "value": "README.md: " + } + }, + "8654b6b67b294522a6c6005820745747": { + "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_f993dc3a96564c188d3cade339ffaf7e", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_4b7bde4ca12d4d3c99d1fd7a3efe1dc9", + "value": 1 + } + }, + "2da39e52fb8d463bae47fdb295fd95c2": { + "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_6ce03d413a0849a78aa4683cbd3333d6", + "placeholder": "​", + "style": "IPY_MODEL_8fe8bf9594dd451d94f976ae77c6c309", + "value": " 5.67k/? [00:00<00:00, 1.52MB/s]" + } + }, + "fab849e9e4d34fdcb47fb6f5b37ce8b7": { + "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 + } + }, + "e71d4fe04ba14e41b3d34012299b33f0": { + "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 + } + }, + "24ea33a9dd494980902f2d2d2d3c2f2e": { + "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": "" + } + }, + "f993dc3a96564c188d3cade339ffaf7e": { + "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": "20px" + } + }, + "4b7bde4ca12d4d3c99d1fd7a3efe1dc9": { + "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": "" + } + }, + "6ce03d413a0849a78aa4683cbd3333d6": { + "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 + } + }, + "8fe8bf9594dd451d94f976ae77c6c309": { + "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": "" + } + }, + "b77c9510ee784585aa2856f8c4e0787f": { + "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_c7e1fe16b5614caf859d30fa44db4be9", + "IPY_MODEL_39773bab82b241fbb2999844c9812516", + "IPY_MODEL_e88a3f9b46a64d2f9f535cd9eb713c70" + ], + "layout": "IPY_MODEL_dcfd19088e03463990d21307614ccd6a" + } + }, + "c7e1fe16b5614caf859d30fa44db4be9": { + "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_441822419f114790abbdea3fe0f87d0d", + "placeholder": "​", + "style": "IPY_MODEL_4197c1a8df1d45999669aba6e088fe83", + "value": "train.jsonl.gz: 100%" + } + }, + "39773bab82b241fbb2999844c9812516": { + "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_2da61cc4d3d24578911ad10a79fd98d0", + "max": 277970, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_cd89f0c06e7c408aa15df2668eb89c2b", + "value": 277970 + } + }, + "e88a3f9b46a64d2f9f535cd9eb713c70": { + "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_5914712138f840d4830a08e1286818bd", + "placeholder": "​", + "style": "IPY_MODEL_8b01634426a646a58a4bb4fff708ea9d", + "value": " 278k/278k [00:00<00:00, 1.39MB/s]" + } + }, + "dcfd19088e03463990d21307614ccd6a": { + "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 + } + }, + "441822419f114790abbdea3fe0f87d0d": { + "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 + } + }, + "4197c1a8df1d45999669aba6e088fe83": { + "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": "" + } + }, + "2da61cc4d3d24578911ad10a79fd98d0": { + "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 + } + }, + "cd89f0c06e7c408aa15df2668eb89c2b": { + "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": "" + } + }, + "5914712138f840d4830a08e1286818bd": { + "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 + } + }, + "8b01634426a646a58a4bb4fff708ea9d": { + "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": "" + } + }, + "a3ec1bca46b54d608ed1cec3e9b86a77": { + "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_e373cfc9515b414893098c943e9b3544", + "IPY_MODEL_7922153899924061b12e434ffb84d37a", + "IPY_MODEL_feb0891d370a4ab0b617e4ff7c9fe7b8" + ], + "layout": "IPY_MODEL_44cb8dc7bbce44c49baf4d5239b6bb25" + } + }, + "e373cfc9515b414893098c943e9b3544": { + "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_33165e1fd85f4d33882db63d0d59cae4", + "placeholder": "​", + "style": "IPY_MODEL_059a83e42aab417a8e3b451d06ecbfe8", + "value": "validation.jsonl.gz: 100%" + } + }, + "7922153899924061b12e434ffb84d37a": { + "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_c9963fa14058479b83ba8fd96addd7d9", + "max": 86431, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7a9f11ff114e4093a2b03366c2273164", + "value": 86431 + } + }, + "feb0891d370a4ab0b617e4ff7c9fe7b8": { + "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_ce79ca6d9e864136be126a1474c8d389", + "placeholder": "​", + "style": "IPY_MODEL_35d1c561a6c145d183f24daf78c71453", + "value": " 86.4k/86.4k [00:00<00:00, 424kB/s]" + } + }, + "44cb8dc7bbce44c49baf4d5239b6bb25": { + "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 + } + }, + "33165e1fd85f4d33882db63d0d59cae4": { + "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 + } + }, + "059a83e42aab417a8e3b451d06ecbfe8": { + "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": "" + } + }, + "c9963fa14058479b83ba8fd96addd7d9": { + "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 + } + }, + "7a9f11ff114e4093a2b03366c2273164": { + "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": "" + } + }, + "ce79ca6d9e864136be126a1474c8d389": { + "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 + } + }, + "35d1c561a6c145d183f24daf78c71453": { + "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": "" + } + }, + "b1f74d52b7594a8db5b1ede3cb901715": { + "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_3e897c0e4f804c1bac03d892bbfe5022", + "IPY_MODEL_2277e5a2cd184f928daeb73c0acbde94", + "IPY_MODEL_43cc92e2c6d245d097ab3d15f3e77f2e" + ], + "layout": "IPY_MODEL_e818f395a11d4fdeaa03a54bddabc91c" + } + }, + "3e897c0e4f804c1bac03d892bbfe5022": { + "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_6f644b61c77b480a868b391ee704d51c", + "placeholder": "​", + "style": "IPY_MODEL_1b35574690a3408eb527b43c6775bbd4", + "value": "test.jsonl.gz: 100%" + } + }, + "2277e5a2cd184f928daeb73c0acbde94": { + "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_90cc3d74316f4c59973a067b035e3d20", + "max": 63205, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0cd89d5a5abc4654965285254ad615da", + "value": 63205 + } + }, + "43cc92e2c6d245d097ab3d15f3e77f2e": { + "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_358642b7ed2a43868cc8b702ce0efc6c", + "placeholder": "​", + "style": "IPY_MODEL_adf03a58a46749b7bae27df5f5b018ed", + "value": " 63.2k/63.2k [00:00<00:00, 310kB/s]" + } + }, + "e818f395a11d4fdeaa03a54bddabc91c": { + "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 + } + }, + "6f644b61c77b480a868b391ee704d51c": { + "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 + } + }, + "1b35574690a3408eb527b43c6775bbd4": { + "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": "" + } + }, + "90cc3d74316f4c59973a067b035e3d20": { + "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 + } + }, + "0cd89d5a5abc4654965285254ad615da": { + "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": "" + } + }, + "358642b7ed2a43868cc8b702ce0efc6c": { + "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 + } + }, + "adf03a58a46749b7bae27df5f5b018ed": { + "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": "" + } + }, + "79f9be2aa6964b42b4eafe00c3856d35": { + "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_8ffcc3129027449b827a5f3edfad279d", + "IPY_MODEL_0a05d9c36747437e9a1962213fdf328e", + "IPY_MODEL_c65bc22e7c654ac2ba342f548ba26344" + ], + "layout": "IPY_MODEL_5f77c2218bfe4af9b27e457c3dd59c9a" + } + }, + "8ffcc3129027449b827a5f3edfad279d": { + "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_282bd748e6874a78ac59f284a9603612", + "placeholder": "​", + "style": "IPY_MODEL_fd695faf735d487f92bb9e22be548f6c", + "value": "Generating train split: 100%" + } + }, + "0a05d9c36747437e9a1962213fdf328e": { + "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_adbda2deb81e4ea68089a699d1e11a4d", + "max": 5749, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_4a06241efad54e4d9ffaf09c8ef1b391", + "value": 5749 + } + }, + "c65bc22e7c654ac2ba342f548ba26344": { + "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_f4b6dcb6d04f42d4b06641f2b3c7fa72", + "placeholder": "​", + "style": "IPY_MODEL_68393608a5d74bd9879581418ef36f4c", + "value": " 5749/5749 [00:00<00:00, 553000.96 examples/s]" + } + }, + "5f77c2218bfe4af9b27e457c3dd59c9a": { + "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 + } + }, + "282bd748e6874a78ac59f284a9603612": { + "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 + } + }, + "fd695faf735d487f92bb9e22be548f6c": { + "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": "" + } + }, + "adbda2deb81e4ea68089a699d1e11a4d": { + "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 + } + }, + "4a06241efad54e4d9ffaf09c8ef1b391": { + "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": "" + } + }, + "f4b6dcb6d04f42d4b06641f2b3c7fa72": { + "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 + } + }, + "68393608a5d74bd9879581418ef36f4c": { + "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": "" + } + }, + "bebc7953dede45b9b483fcebd95d25a7": { + "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_4ee6f2b4b6574d43a3e8eebe8a71e64b", + "IPY_MODEL_2408085f6d0246b690f48e81ddf5918a", + "IPY_MODEL_eeae78b6686a4264b0d220c0682bcb09" + ], + "layout": "IPY_MODEL_eb74e67411924f15ab8d1bef2af9dcce" + } + }, + "4ee6f2b4b6574d43a3e8eebe8a71e64b": { + "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_edb8811d4349428da14d5fc849c5e6d9", + "placeholder": "​", + "style": "IPY_MODEL_e9594b96358a4a88a3ddd737ff7ed646", + "value": "Generating validation split: 100%" + } + }, + "2408085f6d0246b690f48e81ddf5918a": { + "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_11d49bd045244030abde1edf67fd175f", + "max": 1500, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_bf00077cc1e6479595ea23856fb2b857", + "value": 1500 + } + }, + "eeae78b6686a4264b0d220c0682bcb09": { + "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_7790be0e03fc4313bd9a2db205c933fa", + "placeholder": "​", + "style": "IPY_MODEL_b3ca1995e07548608b38e2dc9cd58b4a", + "value": " 1500/1500 [00:00<00:00, 265316.75 examples/s]" + } + }, + "eb74e67411924f15ab8d1bef2af9dcce": { + "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 + } + }, + "edb8811d4349428da14d5fc849c5e6d9": { + "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 + } + }, + "e9594b96358a4a88a3ddd737ff7ed646": { + "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": "" + } + }, + "11d49bd045244030abde1edf67fd175f": { + "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 + } + }, + "bf00077cc1e6479595ea23856fb2b857": { + "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": "" + } + }, + "7790be0e03fc4313bd9a2db205c933fa": { + "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 + } + }, + "b3ca1995e07548608b38e2dc9cd58b4a": { + "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": "" + } + }, + "2db067799aaa458390f2b4f6512b1648": { + "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_c5ea028e856e49499013b6247fbae032", + "IPY_MODEL_0a9708fd8cc3484bac185ce1e3363053", + "IPY_MODEL_8299c7f71a724245aac146136ce68f14" + ], + "layout": "IPY_MODEL_b7e69e58d28842d389ec3af6efc225da" + } + }, + "c5ea028e856e49499013b6247fbae032": { + "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_eaeaab1835084444abc4ee5d175be185", + "placeholder": "​", + "style": "IPY_MODEL_721d0a2d54d44806b465471a40e65e48", + "value": "Generating test split: 100%" + } + }, + "0a9708fd8cc3484bac185ce1e3363053": { + "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_f66edb1067f6413ba8f954a4b1455f53", + "max": 1379, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7ab650307580439aa871ed868dd73e02", + "value": 1379 + } + }, + "8299c7f71a724245aac146136ce68f14": { + "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_b08b134a4a654ce1b58bade830939e43", + "placeholder": "​", + "style": "IPY_MODEL_db5e9dbf211f482fb729859345411acb", + "value": " 1379/1379 [00:00<00:00, 260573.29 examples/s]" + } + }, + "b7e69e58d28842d389ec3af6efc225da": { + "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 + } + }, + "eaeaab1835084444abc4ee5d175be185": { + "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 + } + }, + "721d0a2d54d44806b465471a40e65e48": { + "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": "" + } + }, + "f66edb1067f6413ba8f954a4b1455f53": { + "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 + } + }, + "7ab650307580439aa871ed868dd73e02": { + "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": "" + } + }, + "b08b134a4a654ce1b58bade830939e43": { + "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 + } + }, + "db5e9dbf211f482fb729859345411acb": { + "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": "" + } + }, + "cfc335914cbf43b084f2884bfcd16b08": { + "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_6ddcd1a0106b43898e9bfe457b737b50", + "IPY_MODEL_2a85e704dc5042a783295611891d3673", + "IPY_MODEL_e4d4c1dae6874d54b99f92f89bca7318" + ], + "layout": "IPY_MODEL_f5b69ff3b2a14f4e8b44b012a72fae5b" + } + }, + "6ddcd1a0106b43898e9bfe457b737b50": { + "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_ada3af63f20e46a59ab3d496f477610d", + "placeholder": "​", + "style": "IPY_MODEL_24c77bb630ab4c159a75647090b618de", + "value": "README.md: " + } + }, + "2a85e704dc5042a783295611891d3673": { + "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_20550b2196d64a129e9d45107395ae25", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c6a0987845784bdb8ae012cf8884769a", + "value": 1 + } + }, + "e4d4c1dae6874d54b99f92f89bca7318": { + "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_76fd271429ac46ab8418a8ac6ffc0cb9", + "placeholder": "​", + "style": "IPY_MODEL_764233414e36462f95f94269c63d710b", + "value": " 7.36k/? [00:00<00:00, 2.35MB/s]" + } + }, + "f5b69ff3b2a14f4e8b44b012a72fae5b": { + "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 + } + }, + "ada3af63f20e46a59ab3d496f477610d": { + "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 + } + }, + "24c77bb630ab4c159a75647090b618de": { + "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": "" + } + }, + "20550b2196d64a129e9d45107395ae25": { + "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": "20px" + } + }, + "c6a0987845784bdb8ae012cf8884769a": { + "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": "" + } + }, + "76fd271429ac46ab8418a8ac6ffc0cb9": { + "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 + } + }, + "764233414e36462f95f94269c63d710b": { + "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": "" + } + }, + "189700333aa048728934be7b5d828645": { + "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_30b4705e2bf045f9beac5635eada4e82", + "IPY_MODEL_a414d6711a94476b995ab71e5fb5fccc", + "IPY_MODEL_689bee8e0e964a4a853af9a3e00bb2f0" + ], + "layout": "IPY_MODEL_a52a7381cfdb45d59f2c8d464ecfe4bf" + } + }, + "30b4705e2bf045f9beac5635eada4e82": { + "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_60aa1ad490ec453abac816adbecad28e", + "placeholder": "​", + "style": "IPY_MODEL_0ccfd37c64df4fc2bc9c2019acb4f3fb", + "value": "test.jsonl.gz: 100%" + } + }, + "a414d6711a94476b995ab71e5fb5fccc": { + "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_38b077de9ea74eb6befdca50b10f4a8e", + "max": 174554, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_221b0c6322624ee8817b9c1d35315b73", + "value": 174554 + } + }, + "689bee8e0e964a4a853af9a3e00bb2f0": { + "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_66c3da07e60e4ea4a86739687596620c", + "placeholder": "​", + "style": "IPY_MODEL_dccd8c3c10a34ff9a6de3870ee0645f8", + "value": " 175k/175k [00:00<00:00, 874kB/s]" + } + }, + "a52a7381cfdb45d59f2c8d464ecfe4bf": { + "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 + } + }, + "60aa1ad490ec453abac816adbecad28e": { + "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 + } + }, + "0ccfd37c64df4fc2bc9c2019acb4f3fb": { + "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": "" + } + }, + "38b077de9ea74eb6befdca50b10f4a8e": { + "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 + } + }, + "221b0c6322624ee8817b9c1d35315b73": { + "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": "" + } + }, + "66c3da07e60e4ea4a86739687596620c": { + "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 + } + }, + "dccd8c3c10a34ff9a6de3870ee0645f8": { + "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": "" + } + }, + "e0a5762e6fdd486a897fa94109e3b988": { + "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_d9389ca3df834d2f9a64e8b0bb308a29", + "IPY_MODEL_aeece63f05df4ce78953275473abdbad", + "IPY_MODEL_0b4bc34ed08e4a24b28c280cc3063646" + ], + "layout": "IPY_MODEL_bdd1e56b56cb4079a5c6395c825daf72" + } + }, + "d9389ca3df834d2f9a64e8b0bb308a29": { + "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_c667166f08174334b209b0be75da36f8", + "placeholder": "​", + "style": "IPY_MODEL_7eab67765b894fb1883704e73df5b6d0", + "value": "Generating test split: 100%" + } + }, + "aeece63f05df4ce78953275473abdbad": { + "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_ab9914baaa1a4481b26c702d112b2d5a", + "max": 9927, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c3f947b3292c457f91f8591b35b2564a", + "value": 9927 + } + }, + "0b4bc34ed08e4a24b28c280cc3063646": { + "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_046cf8d05fe444589e58d255dbf5db8f", + "placeholder": "​", + "style": "IPY_MODEL_f5c305fa31ce443c940a9f16c1ef68ae", + "value": " 9927/9927 [00:00<00:00, 1252688.36 examples/s]" + } + }, + "bdd1e56b56cb4079a5c6395c825daf72": { + "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 + } + }, + "c667166f08174334b209b0be75da36f8": { + "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 + } + }, + "7eab67765b894fb1883704e73df5b6d0": { + "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": "" + } + }, + "ab9914baaa1a4481b26c702d112b2d5a": { + "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 + } + }, + "c3f947b3292c457f91f8591b35b2564a": { + "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": "" + } + }, + "046cf8d05fe444589e58d255dbf5db8f": { + "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 + } + }, + "f5c305fa31ce443c940a9f16c1ef68ae": { + "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": "" + } + }, + "508568909bae475b9c29be12d70b0ccd": { + "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_b7bd72832268451f80619571231406e6", + "IPY_MODEL_4e7a916f70064f319c23928c44c1b3de", + "IPY_MODEL_e7e7ee73b9e743f8973fb3ba937d78dc" + ], + "layout": "IPY_MODEL_4462cb5698df46e18cdfccba1f7538f9" + } + }, + "b7bd72832268451f80619571231406e6": { + "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_44b68e0e5de24b5f80043a9429588b66", + "placeholder": "​", + "style": "IPY_MODEL_1c38a351b71e49f580822113040568cc", + "value": "Loading weights: 100%" + } + }, + "4e7a916f70064f319c23928c44c1b3de": { + "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_a11f1eb907c74990a36234b1dc0970ea", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e3b66d14d74048e9ae0d215148121b6b", + "value": 199 + } + }, + "e7e7ee73b9e743f8973fb3ba937d78dc": { + "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_bd5723c7ac0b4211b32d76b8b775f14f", + "placeholder": "​", + "style": "IPY_MODEL_9092d99755a347ac8a45ae016e819ee7", + "value": " 199/199 [00:00<00:00, 4177.09it/s, Materializing param=pooler.dense.weight]" + } + }, + "4462cb5698df46e18cdfccba1f7538f9": { + "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 + } + }, + "44b68e0e5de24b5f80043a9429588b66": { + "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 + } + }, + "1c38a351b71e49f580822113040568cc": { + "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": "" + } + }, + "a11f1eb907c74990a36234b1dc0970ea": { + "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 + } + }, + "e3b66d14d74048e9ae0d215148121b6b": { + "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": "" + } + }, + "bd5723c7ac0b4211b32d76b8b775f14f": { + "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 + } + }, + "9092d99755a347ac8a45ae016e819ee7": { + "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": "" + } + }, + "7f618a7728e948168acd1d46a016022f": { + "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_e9317d14d03d47b58112946964fccba4", + "IPY_MODEL_c3d63defe838429892e07e3fe4743525", + "IPY_MODEL_3c01fb2c7fe947838b39254ef68096bb" + ], + "layout": "IPY_MODEL_bde8ae428f394b2e97af1a09d1171612" + } + }, + "e9317d14d03d47b58112946964fccba4": { + "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_960f59cc25f94d38b47214d9958d9eb0", + "placeholder": "​", + "style": "IPY_MODEL_eb6b2d96790243ee9bc6bd581f1c0d7b", + "value": "Loading weights: 100%" + } + }, + "c3d63defe838429892e07e3fe4743525": { + "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_afb1d381e40e4bc7b96bb3e05bd5b01e", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_79a83cd59ee04936b90d2106330af650", + "value": 134 + } + }, + "3c01fb2c7fe947838b39254ef68096bb": { + "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_e9c2f24d106a43bbaa576b910f7b10a2", + "placeholder": "​", + "style": "IPY_MODEL_8ebfbf0823b44c668ec5db96e177a7f2", + "value": " 134/134 [00:00<00:00, 3963.84it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "bde8ae428f394b2e97af1a09d1171612": { + "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 + } + }, + "960f59cc25f94d38b47214d9958d9eb0": { + "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 + } + }, + "eb6b2d96790243ee9bc6bd581f1c0d7b": { + "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": "" + } + }, + "afb1d381e40e4bc7b96bb3e05bd5b01e": { + "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 + } + }, + "79a83cd59ee04936b90d2106330af650": { + "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": "" + } + }, + "e9c2f24d106a43bbaa576b910f7b10a2": { + "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 + } + }, + "8ebfbf0823b44c668ec5db96e177a7f2": { + "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": "" + } + }, + "925475ff7a5a456690873c59a688d1cf": { + "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_35e71e1fecbb4c5db7d4848056cb273e", + "IPY_MODEL_44ca9db110fa465bb6aea334fae9fb73", + "IPY_MODEL_eb399846160a4e5a9ac1fb44f01c554a" + ], + "layout": "IPY_MODEL_01e5df033e78448c8fea2a10c601eed5" + } + }, + "35e71e1fecbb4c5db7d4848056cb273e": { + "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_8b73d9c5065741cf930497f238fec29c", + "placeholder": "​", + "style": "IPY_MODEL_d2f8cc33ff5b4ed9b183c7a80bfbf17c", + "value": "Loading weights: 100%" + } + }, + "44ca9db110fa465bb6aea334fae9fb73": { + "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_8ab7372bef074e47bc6620bb94949a19", + "max": 197, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5f725231dd86474aa60ac6fde76508b7", + "value": 197 + } + }, + "eb399846160a4e5a9ac1fb44f01c554a": { + "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_93a148c2568943d89dde0d3a9ad9aece", + "placeholder": "​", + "style": "IPY_MODEL_2733ae5d3ae049f0bff7929ee3f256fa", + "value": " 197/197 [00:00<00:00, 4123.12it/s, Materializing param=encoder.layer.11.output.dense.weight]" + } + }, + "01e5df033e78448c8fea2a10c601eed5": { + "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 + } + }, + "8b73d9c5065741cf930497f238fec29c": { + "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 + } + }, + "d2f8cc33ff5b4ed9b183c7a80bfbf17c": { + "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": "" + } + }, + "8ab7372bef074e47bc6620bb94949a19": { + "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 + } + }, + "5f725231dd86474aa60ac6fde76508b7": { + "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": "" + } + }, + "93a148c2568943d89dde0d3a9ad9aece": { + "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 + } + }, + "2733ae5d3ae049f0bff7929ee3f256fa": { + "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": "" + } + }, + "41235a1f501c4280a21a96b6a0415a35": { + "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_fe8ac8b38d5949bf957b0789371af8eb", + "IPY_MODEL_1f6c77432fb74833abecca6146cb2ccf", + "IPY_MODEL_022ed8d5166142b4a5a97d0df26e8048" + ], + "layout": "IPY_MODEL_bfb3b36c2947429cb7474628ed89e573" + } + }, + "fe8ac8b38d5949bf957b0789371af8eb": { + "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_fe6758b05193460bac4f73e05a6c4f5f", + "placeholder": "​", + "style": "IPY_MODEL_ceb23d6435844a3eb4beaab9e103eca6", + "value": "Loading weights: 100%" + } + }, + "1f6c77432fb74833abecca6146cb2ccf": { + "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_256955cf07f04930b76d2c7bcef293ac", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e4afa7b6702b486c8531f2f5c0e1821e", + "value": 25 + } + }, + "022ed8d5166142b4a5a97d0df26e8048": { + "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_ebde16de147546f0918466c31897aef5", + "placeholder": "​", + "style": "IPY_MODEL_d5d72d42cddc48a287b7076758543c95", + "value": " 25/25 [00:00<00:00, 2778.35it/s, Materializing param=pooler.weight]" + } + }, + "bfb3b36c2947429cb7474628ed89e573": { + "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 + } + }, + "fe6758b05193460bac4f73e05a6c4f5f": { + "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 + } + }, + "ceb23d6435844a3eb4beaab9e103eca6": { + "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": "" + } + }, + "256955cf07f04930b76d2c7bcef293ac": { + "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 + } + }, + "e4afa7b6702b486c8531f2f5c0e1821e": { + "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": "" + } + }, + "ebde16de147546f0918466c31897aef5": { + "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 + } + }, + "d5d72d42cddc48a287b7076758543c95": { + "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": "" + } + }, + "6dbac12dff1d4d82a8abf2c295eb7788": { + "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_a4967e28ec7b4203947aff3d3f98ab5f", + "IPY_MODEL_1ab19f3149f341fdb27fb6b2fcd5cbf1", + "IPY_MODEL_b267a673008045d88549e9248bfd3a64" + ], + "layout": "IPY_MODEL_19cfe604fa5c4d3db43f0a32e8a35c42" + } + }, + "a4967e28ec7b4203947aff3d3f98ab5f": { + "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_ab66728ec8eb47b8923f85535a34a8f5", + "placeholder": "​", + "style": "IPY_MODEL_a581ce9c6d684f5d9dca44b18f62a034", + "value": "Loading weights: 100%" + } + }, + "1ab19f3149f341fdb27fb6b2fcd5cbf1": { + "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_7baa5ad8efe44571a33999119a9a4b5d", + "max": 100, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ecbb8b12316c42efb6dec051b55bc1fa", + "value": 100 + } + }, + "b267a673008045d88549e9248bfd3a64": { + "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_ca2a86d355644fdcb204a44ced3d4ec7", + "placeholder": "​", + "style": "IPY_MODEL_d2fa9d75def7493b8878733b386e076c", + "value": " 100/100 [00:00<00:00, 3754.60it/s, Materializing param=transformer.layer.5.sa_layer_norm.weight]" + } + }, + "19cfe604fa5c4d3db43f0a32e8a35c42": { + "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 + } + }, + "ab66728ec8eb47b8923f85535a34a8f5": { + "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 + } + }, + "a581ce9c6d684f5d9dca44b18f62a034": { + "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": "" + } + }, + "7baa5ad8efe44571a33999119a9a4b5d": { + "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 + } + }, + "ecbb8b12316c42efb6dec051b55bc1fa": { + "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": "" + } + }, + "ca2a86d355644fdcb204a44ced3d4ec7": { + "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 + } + }, + "d2fa9d75def7493b8878733b386e076c": { + "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": "" + } + }, + "3dcbb662b30f47cea1f426e9dc400ccd": { + "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_95e7bd361fd64d5da9eba70f738098f1", + "IPY_MODEL_aed55b8a33774489858846e8d91df294", + "IPY_MODEL_cb2850b0ed5c446da78ae62c66f31ab8" + ], + "layout": "IPY_MODEL_177c8cace36648309828ff022da377c6" + } + }, + "95e7bd361fd64d5da9eba70f738098f1": { + "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_18644eff9b28496f8272ea19fc3b7321", + "placeholder": "​", + "style": "IPY_MODEL_f0d53e94e0754085b05a2385bda50e10", + "value": "Loading weights: 100%" + } + }, + "aed55b8a33774489858846e8d91df294": { + "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_407aa645322c43ebb807b1042381e460", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_27987162e66c4fd5be03e00881c1767c", + "value": 134 + } + }, + "cb2850b0ed5c446da78ae62c66f31ab8": { + "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_1502c7cc58e84ee1a9f52e75a9b9b756", + "placeholder": "​", + "style": "IPY_MODEL_b41d4555acca4842bcbcb68fedf14b82", + "value": " 134/134 [00:00<00:00, 4042.35it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "177c8cace36648309828ff022da377c6": { + "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 + } + }, + "18644eff9b28496f8272ea19fc3b7321": { + "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 + } + }, + "f0d53e94e0754085b05a2385bda50e10": { + "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": "" + } + }, + "407aa645322c43ebb807b1042381e460": { + "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 + } + }, + "27987162e66c4fd5be03e00881c1767c": { + "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": "" + } + }, + "1502c7cc58e84ee1a9f52e75a9b9b756": { + "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 + } + }, + "b41d4555acca4842bcbcb68fedf14b82": { + "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": "" + } + }, + "d80a44e073ac4e4d8a080144d253d7fe": { + "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_5e99b7b4abd448f6b52092186dbaa05e", + "IPY_MODEL_37f6bb1b4c384556b7b21667904d38ba", + "IPY_MODEL_0ca39f5cffca45e9b49e99c5547ff5b6" + ], + "layout": "IPY_MODEL_08ff8a06c91646d699bb822e2128e6c4" + } + }, + "5e99b7b4abd448f6b52092186dbaa05e": { + "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_3361780a16294b8aadaf017d28360fd5", + "placeholder": "​", + "style": "IPY_MODEL_8038f4a50f4340489f7c367f33a9dc88", + "value": "Loading weights: 100%" + } + }, + "37f6bb1b4c384556b7b21667904d38ba": { + "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_16f31db48da24f1ea6b6707475a0ba76", + "max": 197, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d8e24943261740ef9b010859cc3fcbc1", + "value": 197 + } + }, + "0ca39f5cffca45e9b49e99c5547ff5b6": { + "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_e72a0689679f4bddbfaad6387d290520", + "placeholder": "​", + "style": "IPY_MODEL_b9dd65320fb24523b8bc0159e0f22eac", + "value": " 197/197 [00:00<00:00, 4261.27it/s, Materializing param=encoder.layer.11.output.dense.weight]" + } + }, + "08ff8a06c91646d699bb822e2128e6c4": { + "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 + } + }, + "3361780a16294b8aadaf017d28360fd5": { + "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 + } + }, + "8038f4a50f4340489f7c367f33a9dc88": { + "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": "" + } + }, + "16f31db48da24f1ea6b6707475a0ba76": { + "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 + } + }, + "d8e24943261740ef9b010859cc3fcbc1": { + "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": "" + } + }, + "e72a0689679f4bddbfaad6387d290520": { + "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 + } + }, + "b9dd65320fb24523b8bc0159e0f22eac": { + "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": "" + } + }, + "d3e519965795424c8e03db777192988f": { + "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_b2178870751745ae8b2ff59efb912a6a", + "IPY_MODEL_596bca946934466ebac67b58bb55a2c6", + "IPY_MODEL_7239062a47784c0fb07e98e430e69c6c" + ], + "layout": "IPY_MODEL_4ce5209809b34b5595c16501acaf36e2" + } + }, + "b2178870751745ae8b2ff59efb912a6a": { + "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_a4038ba6cc2745e9817d5442adb12713", + "placeholder": "​", + "style": "IPY_MODEL_e0dec83560634bd7bac5bbfcaecb4857", + "value": "Loading weights: 100%" + } + }, + "596bca946934466ebac67b58bb55a2c6": { + "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_4261972c71334de188c80f166b653f15", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_10da6bb464ca419eb17496580bf7d88f", + "value": 25 + } + }, + "7239062a47784c0fb07e98e430e69c6c": { + "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_7b8ed631bda04a3dbfaeea4870a2a4b1", + "placeholder": "​", + "style": "IPY_MODEL_9e729736efbd42218c3775334ae2238e", + "value": " 25/25 [00:00<00:00, 2818.60it/s, Materializing param=pooler.weight]" + } + }, + "4ce5209809b34b5595c16501acaf36e2": { + "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 + } + }, + "a4038ba6cc2745e9817d5442adb12713": { + "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 + } + }, + "e0dec83560634bd7bac5bbfcaecb4857": { + "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": "" + } + }, + "4261972c71334de188c80f166b653f15": { + "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 + } + }, + "10da6bb464ca419eb17496580bf7d88f": { + "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": "" + } + }, + "7b8ed631bda04a3dbfaeea4870a2a4b1": { + "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 + } + }, + "9e729736efbd42218c3775334ae2238e": { + "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": "" + } + }, + "db8fb5f7d5d641e0a63c27883875fbb8": { + "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_359bf897b2404f4ca3b1250d8d060c8b", + "IPY_MODEL_21233148c885454d9d1ccc5467b0877c", + "IPY_MODEL_b8a4c7d2168743028374e094fdba3778" + ], + "layout": "IPY_MODEL_e001c3cb8a1a44cca95d5d191749e596" + } + }, + "359bf897b2404f4ca3b1250d8d060c8b": { + "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_49fb49ff55c14127b11e942667fddc5a", + "placeholder": "​", + "style": "IPY_MODEL_251a783db5f74b50a7b32970c85fac1a", + "value": "Loading weights: 100%" + } + }, + "21233148c885454d9d1ccc5467b0877c": { + "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_bb23409acd204b52b03879633a2a3f81", + "max": 100, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_684d665814904e2ba054607b97a57b55", + "value": 100 + } + }, + "b8a4c7d2168743028374e094fdba3778": { + "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_ff5545a913c140ff96f42e3334942bdc", + "placeholder": "​", + "style": "IPY_MODEL_a780ba2ef7834be6882edccace4cbe3b", + "value": " 100/100 [00:00<00:00, 3620.90it/s, Materializing param=transformer.layer.5.sa_layer_norm.weight]" + } + }, + "e001c3cb8a1a44cca95d5d191749e596": { + "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 + } + }, + "49fb49ff55c14127b11e942667fddc5a": { + "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 + } + }, + "251a783db5f74b50a7b32970c85fac1a": { + "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": "" + } + }, + "bb23409acd204b52b03879633a2a3f81": { + "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 + } + }, + "684d665814904e2ba054607b97a57b55": { + "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": "" + } + }, + "ff5545a913c140ff96f42e3334942bdc": { + "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 + } + }, + "a780ba2ef7834be6882edccace4cbe3b": { + "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": "" + } + }, + "695cc44790d54dc197cbfc06c73de26c": { + "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_65bbbfa5989e4e69856f486ffeb51b40", + "IPY_MODEL_fe6306f65bd64dff8effe42a6fb349f7", + "IPY_MODEL_f4d7984d22204ad69d2b55af4c91f794" + ], + "layout": "IPY_MODEL_6ef8994ad8274c46b9ee92e573280809" + } + }, + "65bbbfa5989e4e69856f486ffeb51b40": { + "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_2efbd594188d41169cb3b6b24288006f", + "placeholder": "​", + "style": "IPY_MODEL_04d91c3b2564478a9f0bfe10a311d7ec", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "fe6306f65bd64dff8effe42a6fb349f7": { + "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_1bcb26c4ab5f47119b129545e977ce47", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_2e42ec41bd354de3a201f22a9a7c593d", + "value": 1 + } + }, + "f4d7984d22204ad69d2b55af4c91f794": { + "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_9468f62eeb4d48ac881e3fc106b9fdf5", + "placeholder": "​", + "style": "IPY_MODEL_a1bdda291b7b4eca83dfbfd670321264", + "value": "  104MB /  104MB, 16.8MB/s  " + } + }, + "6ef8994ad8274c46b9ee92e573280809": { + "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 + } + }, + "2efbd594188d41169cb3b6b24288006f": { + "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 + } + }, + "04d91c3b2564478a9f0bfe10a311d7ec": { + "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": "" + } + }, + "1bcb26c4ab5f47119b129545e977ce47": { + "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": "20px" + } + }, + "2e42ec41bd354de3a201f22a9a7c593d": { + "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": "" + } + }, + "9468f62eeb4d48ac881e3fc106b9fdf5": { + "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 + } + }, + "a1bdda291b7b4eca83dfbfd670321264": { + "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": "" + } + }, + "1d12abb148e5454aae18dcdefaf7bdf6": { + "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_1b964e9e63754d25abce781a49355638", + "IPY_MODEL_814735e5616f422d8ff9bbeee352f827", + "IPY_MODEL_a51b8488b9d941ffa8139fcc0247e9a6" + ], + "layout": "IPY_MODEL_5fd166b88db9486190bc9b6329c1b98b" + } + }, + "1b964e9e63754d25abce781a49355638": { + "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_08cca774026641eab9e2da04c916a2eb", + "placeholder": "​", + "style": "IPY_MODEL_680808a7d7094638811eebab75a20f7f", + "value": "New Data Upload               : 100%" + } + }, + "814735e5616f422d8ff9bbeee352f827": { + "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_5b37a0473b504dfe854408d095e93fc4", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_139085515c264c7491b9cfb10f996ca8", + "value": 1 + } + }, + "a51b8488b9d941ffa8139fcc0247e9a6": { + "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_521fa2c3df1f4cab9c97ac9d5f6868dd", + "placeholder": "​", + "style": "IPY_MODEL_e67d6854b61b47568ac0cb11fce23906", + "value": "  104MB /  104MB, 16.8MB/s  " + } + }, + "5fd166b88db9486190bc9b6329c1b98b": { + "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 + } + }, + "08cca774026641eab9e2da04c916a2eb": { + "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 + } + }, + "680808a7d7094638811eebab75a20f7f": { + "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": "" + } + }, + "5b37a0473b504dfe854408d095e93fc4": { + "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": "20px" + } + }, + "139085515c264c7491b9cfb10f996ca8": { + "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": "" + } + }, + "521fa2c3df1f4cab9c97ac9d5f6868dd": { + "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 + } + }, + "e67d6854b61b47568ac0cb11fce23906": { + "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": "" + } + }, + "de12ccc125be4d63b32ddd65c9b7d59a": { + "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_dc7a5a39f30248dc856c65030b7e059d", + "IPY_MODEL_85dabc8a64304f329e79c069b90fe1cc", + "IPY_MODEL_51f2203949b340eeba16d2826b94d594" + ], + "layout": "IPY_MODEL_53d80988fc5d48fdbff27feecb805162" + } + }, + "dc7a5a39f30248dc856c65030b7e059d": { + "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_b528c9750e2c407fb05c6b1b99cd64ba", + "placeholder": "​", + "style": "IPY_MODEL_c79ca2e80bd14825b93230c553d21798", + "value": "  ...00k/student/best_model.pt: 100%" + } + }, + "85dabc8a64304f329e79c069b90fe1cc": { + "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_d82ef80e885d4d528b30e39983c5357e", + "max": 103862453, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b152a746782041eaa9cd74183a19b760", + "value": 103862453 + } + }, + "51f2203949b340eeba16d2826b94d594": { + "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_3f735587627c4463913ede49a113a0f1", + "placeholder": "​", + "style": "IPY_MODEL_bc0a31b1d38e49a5813ee18a3e8562de", + "value": "  104MB /  104MB            " + } + }, + "53d80988fc5d48fdbff27feecb805162": { + "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 + } + }, + "b528c9750e2c407fb05c6b1b99cd64ba": { + "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 + } + }, + "c79ca2e80bd14825b93230c553d21798": { + "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": "" + } + }, + "d82ef80e885d4d528b30e39983c5357e": { + "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 + } + }, + "b152a746782041eaa9cd74183a19b760": { + "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": "" + } + }, + "3f735587627c4463913ede49a113a0f1": { + "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 + } + }, + "bc0a31b1d38e49a5813ee18a3e8562de": { + "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": "" + } + }, + "eca8bdbc9b5e4df4bdbcd7ae8271e50d": { + "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_d6b61777bad149a4962c331addefd275", + "IPY_MODEL_292bd24f0e0a45ea8ff43f7ca11d033e", + "IPY_MODEL_8673d506828a42a09b5a8b1cde0bd8fa" + ], + "layout": "IPY_MODEL_a012e21653ab479fa9f72c2e1ee93471" + } + }, + "d6b61777bad149a4962c331addefd275": { + "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_45d571bf4daa4c7e9b1e56b76f800695", + "placeholder": "​", + "style": "IPY_MODEL_964a036f57db4659815d6447343ed026", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "292bd24f0e0a45ea8ff43f7ca11d033e": { + "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_298d63cae3c1400faba811f7a3cbe5fe", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_72d4cffa500b4b2e8e59a9463b1ece98", + "value": 1 + } + }, + "8673d506828a42a09b5a8b1cde0bd8fa": { + "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_0c2f2872f4e14b05b1839da60700eb5b", + "placeholder": "​", + "style": "IPY_MODEL_bcefbafa83b5449ea5e90f00c7a84bca", + "value": "  104MB /  104MB, 31.6MB/s  " + } + }, + "a012e21653ab479fa9f72c2e1ee93471": { + "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 + } + }, + "45d571bf4daa4c7e9b1e56b76f800695": { + "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 + } + }, + "964a036f57db4659815d6447343ed026": { + "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": "" + } + }, + "298d63cae3c1400faba811f7a3cbe5fe": { + "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": "20px" + } + }, + "72d4cffa500b4b2e8e59a9463b1ece98": { + "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": "" + } + }, + "0c2f2872f4e14b05b1839da60700eb5b": { + "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 + } + }, + "bcefbafa83b5449ea5e90f00c7a84bca": { + "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": "" + } + }, + "963b4f0aa6734a2993d053e6703f31b2": { + "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_cfd4f6aaafc14bc3a794a5a5278c3ca1", + "IPY_MODEL_8a5204c97695413daf5b013b8edb7d7b", + "IPY_MODEL_ca84f9dea6c746e6afa89a33f414bb3d" + ], + "layout": "IPY_MODEL_c03ffe5b5d48492c895f6e6978a6f228" + } + }, + "cfd4f6aaafc14bc3a794a5a5278c3ca1": { + "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_a23581d25fbd438fabaf42a3074e6a5e", + "placeholder": "​", + "style": "IPY_MODEL_a3749428577e438d8447dd5f2315930f", + "value": "New Data Upload               : 100%" + } + }, + "8a5204c97695413daf5b013b8edb7d7b": { + "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_5687f629be5944f28465dae51d8ec326", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_497b634247d84594ac2a9f07c72547f3", + "value": 1 + } + }, + "ca84f9dea6c746e6afa89a33f414bb3d": { + "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_812422d191b846278efe823f539f95ee", + "placeholder": "​", + "style": "IPY_MODEL_9dea8d1a7ce94133b0b2f23b9745a7ac", + "value": " 2.75MB / 2.75MB, 3.44MB/s  " + } + }, + "c03ffe5b5d48492c895f6e6978a6f228": { + "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 + } + }, + "a23581d25fbd438fabaf42a3074e6a5e": { + "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 + } + }, + "a3749428577e438d8447dd5f2315930f": { + "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": "" + } + }, + "5687f629be5944f28465dae51d8ec326": { + "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": "20px" + } + }, + "497b634247d84594ac2a9f07c72547f3": { + "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": "" + } + }, + "812422d191b846278efe823f539f95ee": { + "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 + } + }, + "9dea8d1a7ce94133b0b2f23b9745a7ac": { + "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": "" + } + }, + "c3556e3cf3b64791a703987c8b6c5ab2": { + "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_ed9757a6b9ee4a6e99f9813410d4be67", + "IPY_MODEL_e531a7b0cacc47b0a6a98339185faef8", + "IPY_MODEL_c29a557c061b45a09e58643573cd38b0" + ], + "layout": "IPY_MODEL_5974914eeb5c4cee80c5f5a39c310381" + } + }, + "ed9757a6b9ee4a6e99f9813410d4be67": { + "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_a2e21df86bf241de95b5539e500ee24b", + "placeholder": "​", + "style": "IPY_MODEL_67756821a5a448f796440ef2e4869fe2", + "value": "  ...0k/student/final_model.pt: 100%" + } + }, + "e531a7b0cacc47b0a6a98339185faef8": { + "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_563e1ca0d17642f38fd6f1ab18960edd", + "max": 103862541, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_1f1952acdb394a6a953543019c1b74d0", + "value": 103862541 + } + }, + "c29a557c061b45a09e58643573cd38b0": { + "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_b3d268a8d6e9497c9a40cc70b8b9c007", + "placeholder": "​", + "style": "IPY_MODEL_ed470401cd8a4e878b6f5142b092e965", + "value": "  104MB /  104MB            " + } + }, + "5974914eeb5c4cee80c5f5a39c310381": { + "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 + } + }, + "a2e21df86bf241de95b5539e500ee24b": { + "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 + } + }, + "67756821a5a448f796440ef2e4869fe2": { + "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": "" + } + }, + "563e1ca0d17642f38fd6f1ab18960edd": { + "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 + } + }, + "1f1952acdb394a6a953543019c1b74d0": { + "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": "" + } + }, + "b3d268a8d6e9497c9a40cc70b8b9c007": { + "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 + } + }, + "ed470401cd8a4e878b6f5142b092e965": { + "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": "" + } + }, + "ab19b4e6d2bd4c708d5894cd3fe2bf76": { + "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_fd3fd1e3507745fa8165748489455a95", + "IPY_MODEL_089231ae267a4bb280c6a475c55e7aa7", + "IPY_MODEL_5e266791178c44c2ae521d0ad2d0fb77" + ], + "layout": "IPY_MODEL_5195cca927f34cb697f931ca0919f423" + } + }, + "fd3fd1e3507745fa8165748489455a95": { + "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_a328b378f9e34b6099079195a204ae16", + "placeholder": "​", + "style": "IPY_MODEL_ea0918b3f57a4dcf9d4b057e5c7aefd0", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "089231ae267a4bb280c6a475c55e7aa7": { + "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_87b0e8fb5025487da07e607adbadaf37", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c044bcd8f3814eecae8b9deebe51cbee", + "value": 1 + } + }, + "5e266791178c44c2ae521d0ad2d0fb77": { + "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_130f0e7cc2dc4e099f5b6424a78b10a0", + "placeholder": "​", + "style": "IPY_MODEL_30c704bfe00144ffb1890dde7edf8b83", + "value": "  104MB /  104MB, 22.6MB/s  " + } + }, + "5195cca927f34cb697f931ca0919f423": { + "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 + } + }, + "a328b378f9e34b6099079195a204ae16": { + "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 + } + }, + "ea0918b3f57a4dcf9d4b057e5c7aefd0": { + "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": "" + } + }, + "87b0e8fb5025487da07e607adbadaf37": { + "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": "20px" + } + }, + "c044bcd8f3814eecae8b9deebe51cbee": { + "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": "" + } + }, + "130f0e7cc2dc4e099f5b6424a78b10a0": { + "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 + } + }, + "30c704bfe00144ffb1890dde7edf8b83": { + "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": "" + } + }, + "9d047fc748274b218837d247a5531812": { + "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_658b4a031b7e4c57b0e61133913353bd", + "IPY_MODEL_4ec10750717343928ef6f6e252b2bbdd", + "IPY_MODEL_0a704159a5214586b3e117fc5e9c0984" + ], + "layout": "IPY_MODEL_b51213c349c34cb0890ade77f3bb14bb" + } + }, + "658b4a031b7e4c57b0e61133913353bd": { + "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_2c3d39168a8243f5a750a9088c870944", + "placeholder": "​", + "style": "IPY_MODEL_c5eb5d2dc79446bd998bfdcd11447f52", + "value": "New Data Upload               : 100%" + } + }, + "4ec10750717343928ef6f6e252b2bbdd": { + "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_3d9a4be53ad94248ae3ded952c94720f", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d02aa3bbc5624e24a57989bace413251", + "value": 1 + } + }, + "0a704159a5214586b3e117fc5e9c0984": { + "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_f0f14ed0e855474698e1ad1f921bf5f4", + "placeholder": "​", + "style": "IPY_MODEL_eb77b6f3c6c24152b7e93e5aaa4df41e", + "value": "  104MB /  104MB, 22.6MB/s  " + } + }, + "b51213c349c34cb0890ade77f3bb14bb": { + "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 + } + }, + "2c3d39168a8243f5a750a9088c870944": { + "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 + } + }, + "c5eb5d2dc79446bd998bfdcd11447f52": { + "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": "" + } + }, + "3d9a4be53ad94248ae3ded952c94720f": { + "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": "20px" + } + }, + "d02aa3bbc5624e24a57989bace413251": { + "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": "" + } + }, + "f0f14ed0e855474698e1ad1f921bf5f4": { + "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 + } + }, + "eb77b6f3c6c24152b7e93e5aaa4df41e": { + "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": "" + } + }, + "b2e51952cf234c289166a89725c0b838": { + "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_0d222d0fa3a646a88e9ac251f43fc1af", + "IPY_MODEL_074f4e4bf88f45f083be4ff38d145cc0", + "IPY_MODEL_a06972e6f6ae48a994cf66240a2651aa" + ], + "layout": "IPY_MODEL_da598c9ad10f49c5afe2673150359118" + } + }, + "0d222d0fa3a646a88e9ac251f43fc1af": { + "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_2f7b02616aa94963b0c643f4a8c24978", + "placeholder": "​", + "style": "IPY_MODEL_878137c2dd3c4d6b882849bf054a4438", + "value": "  ...500k/student/model_e10.pt: 100%" + } + }, + "074f4e4bf88f45f083be4ff38d145cc0": { + "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_50189050588940178cb52391f84b6dc4", + "max": 103862365, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_37814d2464ef4d6f84e4d3f6ab995ecc", + "value": 103862365 + } + }, + "a06972e6f6ae48a994cf66240a2651aa": { + "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_47433bde0081459e8e94cf8485e22267", + "placeholder": "​", + "style": "IPY_MODEL_f5b19dc02b014e34b7a99c9c037e092c", + "value": "  104MB /  104MB            " + } + }, + "da598c9ad10f49c5afe2673150359118": { + "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 + } + }, + "2f7b02616aa94963b0c643f4a8c24978": { + "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 + } + }, + "878137c2dd3c4d6b882849bf054a4438": { + "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": "" + } + }, + "50189050588940178cb52391f84b6dc4": { + "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 + } + }, + "37814d2464ef4d6f84e4d3f6ab995ecc": { + "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": "" + } + }, + "47433bde0081459e8e94cf8485e22267": { + "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 + } + }, + "f5b19dc02b014e34b7a99c9c037e092c": { + "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": "" + } + }, + "b209e3de1ee34fbe8e9180c47646d822": { + "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_c2357c5b48494957a34c8d354d532080", + "IPY_MODEL_b5c2eca08f314672991fc1d15e2540ae", + "IPY_MODEL_66cb4f41b950475281349fe04f0e98c3" + ], + "layout": "IPY_MODEL_4c05f90084fd4e3990f53ecef8107433" + } + }, + "c2357c5b48494957a34c8d354d532080": { + "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_171383fd4fc3484da9cc1d1dbf24f76f", + "placeholder": "​", + "style": "IPY_MODEL_f0536a6fad8e4d4194e016461b7189d4", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "b5c2eca08f314672991fc1d15e2540ae": { + "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_a4526b5e2bce45939e4a4ca163001b8e", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_a24cae9e39a4469caf557a1604c44e57", + "value": 1 + } + }, + "66cb4f41b950475281349fe04f0e98c3": { + "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_77f4ba6fc5fc4bea86845ca62180afa2", + "placeholder": "​", + "style": "IPY_MODEL_e30f8054cd7149b3ba5cec5bac09d62a", + "value": "  104MB /  104MB, 24.7MB/s  " + } + }, + "4c05f90084fd4e3990f53ecef8107433": { + "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 + } + }, + "171383fd4fc3484da9cc1d1dbf24f76f": { + "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 + } + }, + "f0536a6fad8e4d4194e016461b7189d4": { + "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": "" + } + }, + "a4526b5e2bce45939e4a4ca163001b8e": { + "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": "20px" + } + }, + "a24cae9e39a4469caf557a1604c44e57": { + "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": "" + } + }, + "77f4ba6fc5fc4bea86845ca62180afa2": { + "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 + } + }, + "e30f8054cd7149b3ba5cec5bac09d62a": { + "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": "" + } + }, + "06175ec4ef08403aa1a574b64c637eb1": { + "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_91d0651c23cb4a2eab6c89f394d0a8b6", + "IPY_MODEL_49ea0246b4f0438baa977068e19d7c27", + "IPY_MODEL_508f71146b11460a858d75f9052cd77e" + ], + "layout": "IPY_MODEL_5af51489cd1c43e8be1f976392c6aaab" + } + }, + "91d0651c23cb4a2eab6c89f394d0a8b6": { + "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_671436edc37a491ebd18a40aacd59b00", + "placeholder": "​", + "style": "IPY_MODEL_caa1905421104cb99f9acb28d6fe605a", + "value": "New Data Upload               : 100%" + } + }, + "49ea0246b4f0438baa977068e19d7c27": { + "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_de5c9bff0fe3415b9a00bbb0194fda44", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_2f4ce1db3fee457397a848a44a9432ae", + "value": 1 + } + }, + "508f71146b11460a858d75f9052cd77e": { + "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_8500d0c4788440a49e0e884a9b02591c", + "placeholder": "​", + "style": "IPY_MODEL_5ac0b986d9484dcf8051e3f69a36ebd2", + "value": "  104MB /  104MB, 24.7MB/s  " + } + }, + "5af51489cd1c43e8be1f976392c6aaab": { + "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 + } + }, + "671436edc37a491ebd18a40aacd59b00": { + "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 + } + }, + "caa1905421104cb99f9acb28d6fe605a": { + "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": "" + } + }, + "de5c9bff0fe3415b9a00bbb0194fda44": { + "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": "20px" + } + }, + "2f4ce1db3fee457397a848a44a9432ae": { + "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": "" + } + }, + "8500d0c4788440a49e0e884a9b02591c": { + "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 + } + }, + "5ac0b986d9484dcf8051e3f69a36ebd2": { + "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": "" + } + }, + "ed0122a1e258458fa8b22d7ad9c41f84": { + "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_3d10efe14ca241f798b065cda005bc84", + "IPY_MODEL_06f8917e443d42058c0a19946dd22130", + "IPY_MODEL_154dfcdd62104c758ea3603474283508" + ], + "layout": "IPY_MODEL_94d1827c3bb34bf58563916466e982d6" + } + }, + "3d10efe14ca241f798b065cda005bc84": { + "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_eee4e90db7f64c0b8ba3f4e23b35b16a", + "placeholder": "​", + "style": "IPY_MODEL_e1ce8464bf6846fc8035933566f79c6b", + "value": "  ...500k/student/model_e20.pt: 100%" + } + }, + "06f8917e443d42058c0a19946dd22130": { + "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_2622826856b24a8dbd366d2b6c5861cb", + "max": 103862365, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_de1595a123e34af0a8e0a82cf2147694", + "value": 103862365 + } + }, + "154dfcdd62104c758ea3603474283508": { + "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_718edaf01ba845c78bc042d0baa90244", + "placeholder": "​", + "style": "IPY_MODEL_898880c4cf5c4dbeaaf23d139aab1bfd", + "value": "  104MB /  104MB            " + } + }, + "94d1827c3bb34bf58563916466e982d6": { + "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 + } + }, + "eee4e90db7f64c0b8ba3f4e23b35b16a": { + "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 + } + }, + "e1ce8464bf6846fc8035933566f79c6b": { + "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": "" + } + }, + "2622826856b24a8dbd366d2b6c5861cb": { + "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 + } + }, + "de1595a123e34af0a8e0a82cf2147694": { + "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": "" + } + }, + "718edaf01ba845c78bc042d0baa90244": { + "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 + } + }, + "898880c4cf5c4dbeaaf23d139aab1bfd": { + "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": "" + } + }, + "6d1bed0ff54040b8a2b4450111690301": { + "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_de6391b734a142bd8dc5bba52b95cbec", + "IPY_MODEL_003e45edf8ca4806b5def8d226b26977", + "IPY_MODEL_277faa5a00fe48b58a9aec3bf7380589" + ], + "layout": "IPY_MODEL_5dc60147aed14fa3843e17d60ab64558" + } + }, + "de6391b734a142bd8dc5bba52b95cbec": { + "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_070bf7752f16497e8d38522affec4bda", + "placeholder": "​", + "style": "IPY_MODEL_d4570a81755d449e9902f91d98e40459", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "003e45edf8ca4806b5def8d226b26977": { + "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_3b7b9a47a4e64bdabb523b5877e1017f", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_69d6398161bb40799abd34e377cecaf6", + "value": 1 + } + }, + "277faa5a00fe48b58a9aec3bf7380589": { + "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_4a7db03ebf514940979812cd3c59b68b", + "placeholder": "​", + "style": "IPY_MODEL_6f501567ecab4a4aaa814f50682f7a99", + "value": "  104MB /  104MB, 31.5MB/s  " + } + }, + "5dc60147aed14fa3843e17d60ab64558": { + "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 + } + }, + "070bf7752f16497e8d38522affec4bda": { + "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 + } + }, + "d4570a81755d449e9902f91d98e40459": { + "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": "" + } + }, + "3b7b9a47a4e64bdabb523b5877e1017f": { + "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": "20px" + } + }, + "69d6398161bb40799abd34e377cecaf6": { + "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": "" + } + }, + "4a7db03ebf514940979812cd3c59b68b": { + "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 + } + }, + "6f501567ecab4a4aaa814f50682f7a99": { + "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": "" + } + }, + "98800a5db73945f8abe5031df0654564": { + "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_cd8fef2b8b7142e9821d9577042ad4f3", + "IPY_MODEL_ed17fcf6cd6a464b9e125c081d6de5bd", + "IPY_MODEL_98f2570ccd254719ba63a1695b43b9b8" + ], + "layout": "IPY_MODEL_4e8698d3ec69485c840ef7928ad83290" + } + }, + "cd8fef2b8b7142e9821d9577042ad4f3": { + "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_a7ecb0c07a1d44728393dbac548809b0", + "placeholder": "​", + "style": "IPY_MODEL_4be1cff254b44a22b429fd56fc270f6f", + "value": "New Data Upload               : 100%" + } + }, + "ed17fcf6cd6a464b9e125c081d6de5bd": { + "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_4a72d8ebd2844d68b07527ce816e8a04", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_f8251ae7a766421fba7520cad1959cc7", + "value": 1 + } + }, + "98f2570ccd254719ba63a1695b43b9b8": { + "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_a44cf930d17d4a80ac21bb5808e3b0f6", + "placeholder": "​", + "style": "IPY_MODEL_d096f7eebdff4f378c0ffef6ba1f688c", + "value": " 2.75MB / 2.75MB, 3.44MB/s  " + } + }, + "4e8698d3ec69485c840ef7928ad83290": { + "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 + } + }, + "a7ecb0c07a1d44728393dbac548809b0": { + "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 + } + }, + "4be1cff254b44a22b429fd56fc270f6f": { + "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": "" + } + }, + "4a72d8ebd2844d68b07527ce816e8a04": { + "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": "20px" + } + }, + "f8251ae7a766421fba7520cad1959cc7": { + "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": "" + } + }, + "a44cf930d17d4a80ac21bb5808e3b0f6": { + "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 + } + }, + "d096f7eebdff4f378c0ffef6ba1f688c": { + "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": "" + } + }, + "eea186c0522a4a3c9884abe53986c7a4": { + "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_84b82b489abe4037b4e3c1493caa86d7", + "IPY_MODEL_155079894a68437b8791bd1b42e6e419", + "IPY_MODEL_d2dd137b952348b0996be0c67a1a112c" + ], + "layout": "IPY_MODEL_191ae267f8614e3993c7627ce927ae6e" + } + }, + "84b82b489abe4037b4e3c1493caa86d7": { + "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_66ee8aa450a94bbca05e5dd70a3890cb", + "placeholder": "​", + "style": "IPY_MODEL_9c2ebe4e5f514128b401d9c4f5359a93", + "value": "  ...500k/student/model_e30.pt: 100%" + } + }, + "155079894a68437b8791bd1b42e6e419": { + "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_2feffee0d1aa4119b87998bb3cff9663", + "max": 103862365, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0dd584049b304c87b33cf380b4e53cd3", + "value": 103862365 + } + }, + "d2dd137b952348b0996be0c67a1a112c": { + "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_802f47a40fb447ec9c123d4ecd7581cb", + "placeholder": "​", + "style": "IPY_MODEL_592fa62bb6ef48f8a5e43f791ef39e40", + "value": "  104MB /  104MB            " + } + }, + "191ae267f8614e3993c7627ce927ae6e": { + "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 + } + }, + "66ee8aa450a94bbca05e5dd70a3890cb": { + "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 + } + }, + "9c2ebe4e5f514128b401d9c4f5359a93": { + "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": "" + } + }, + "2feffee0d1aa4119b87998bb3cff9663": { + "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 + } + }, + "0dd584049b304c87b33cf380b4e53cd3": { + "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": "" + } + }, + "802f47a40fb447ec9c123d4ecd7581cb": { + "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 + } + }, + "592fa62bb6ef48f8a5e43f791ef39e40": { + "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": "" + } + }, + "8a0cd81be2be47888596acf176afddb9": { + "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_84f657f86bd045359a1a1759d7194f76", + "IPY_MODEL_dec09fc897a84c4189fdf2c4d26bc643", + "IPY_MODEL_b02c30d0e2e44d9eb97699ea04616bc1" + ], + "layout": "IPY_MODEL_58509995a438455ba23ebe170493bc7e" + } + }, + "84f657f86bd045359a1a1759d7194f76": { + "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_3c06e0278e234c77ac77a750c2c5d578", + "placeholder": "​", + "style": "IPY_MODEL_fb7c3e3c8942451baf4ca758e9a6a946", + "value": "Loading weights: 100%" + } + }, + "dec09fc897a84c4189fdf2c4d26bc643": { + "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_d7e01fe40f6c4501875a99052345c3f0", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5cb8b3e773b84a85ac577226e2857edc", + "value": 199 + } + }, + "b02c30d0e2e44d9eb97699ea04616bc1": { + "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_4dd803a5e553472cbf00521945568d80", + "placeholder": "​", + "style": "IPY_MODEL_fff44aa0989441a596aa08b7bdf0b6b1", + "value": " 199/199 [00:00<00:00, 4149.43it/s, Materializing param=pooler.dense.weight]" + } + }, + "58509995a438455ba23ebe170493bc7e": { + "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 + } + }, + "3c06e0278e234c77ac77a750c2c5d578": { + "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 + } + }, + "fb7c3e3c8942451baf4ca758e9a6a946": { + "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": "" + } + }, + "d7e01fe40f6c4501875a99052345c3f0": { + "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 + } + }, + "5cb8b3e773b84a85ac577226e2857edc": { + "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": "" + } + }, + "4dd803a5e553472cbf00521945568d80": { + "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 + } + }, + "fff44aa0989441a596aa08b7bdf0b6b1": { + "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": "" + } + }, + "e99c0c2fd25c4688bde03694ea6fdc19": { + "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_8871d95b022b415693cb2db9a3e7b775", + "IPY_MODEL_a64c92181f864e41a9637e4ffd3cdb1c", + "IPY_MODEL_30cdef2f92ba44eb9b42f442af3b6c42" + ], + "layout": "IPY_MODEL_6f9d36f1066b43d1986ce17cbfc95d9c" + } + }, + "8871d95b022b415693cb2db9a3e7b775": { + "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_945d01f31faf4f929acfe4e99f9bde98", + "placeholder": "​", + "style": "IPY_MODEL_f4c52fb0719e4d38bbc1bc680117be78", + "value": "Loading weights: 100%" + } + }, + "a64c92181f864e41a9637e4ffd3cdb1c": { + "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_2b1cf94985fc4440a22cc337f71f2f9c", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_122b374495c54486b721dea6f2ecf49a", + "value": 134 + } + }, + "30cdef2f92ba44eb9b42f442af3b6c42": { + "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_0bacd6dcbe5f407dad12ff88ca0c734d", + "placeholder": "​", + "style": "IPY_MODEL_8e8d48014e514158ac40d1ca480dbe84", + "value": " 134/134 [00:00<00:00, 4108.46it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "6f9d36f1066b43d1986ce17cbfc95d9c": { + "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 + } + }, + "945d01f31faf4f929acfe4e99f9bde98": { + "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 + } + }, + "f4c52fb0719e4d38bbc1bc680117be78": { + "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": "" + } + }, + "2b1cf94985fc4440a22cc337f71f2f9c": { + "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 + } + }, + "122b374495c54486b721dea6f2ecf49a": { + "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": "" + } + }, + "0bacd6dcbe5f407dad12ff88ca0c734d": { + "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 + } + }, + "8e8d48014e514158ac40d1ca480dbe84": { + "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": "" + } + }, + "0e9d33c2824249bc9543659f9810e469": { + "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_a02595570db14b8c949550d035f79d9e", + "IPY_MODEL_17c946ace59444ec9b8504bcc96e7bb9", + "IPY_MODEL_892fc7ea327d4bd29228cb577cf709ba" + ], + "layout": "IPY_MODEL_2646ff0b9bb941b3aeff743bc3c23ea9" + } + }, + "a02595570db14b8c949550d035f79d9e": { + "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_0690fff8b38f4a44a07ad90232e7e727", + "placeholder": "​", + "style": "IPY_MODEL_003b40cfdf664f76958b2802917db0a3", + "value": "Loading weights: 100%" + } + }, + "17c946ace59444ec9b8504bcc96e7bb9": { + "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_a56949abd7b24bee8eb81586738815d7", + "max": 197, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_748256ef83b94c0a9869503223dd2fcc", + "value": 197 + } + }, + "892fc7ea327d4bd29228cb577cf709ba": { + "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_f2152636947d427bb30f8c8a77247dd8", + "placeholder": "​", + "style": "IPY_MODEL_6693c870f6494f40a3b64b021960d2b0", + "value": " 197/197 [00:00<00:00, 4224.84it/s, Materializing param=encoder.layer.11.output.dense.weight]" + } + }, + "2646ff0b9bb941b3aeff743bc3c23ea9": { + "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 + } + }, + "0690fff8b38f4a44a07ad90232e7e727": { + "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 + } + }, + "003b40cfdf664f76958b2802917db0a3": { + "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": "" + } + }, + "a56949abd7b24bee8eb81586738815d7": { + "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 + } + }, + "748256ef83b94c0a9869503223dd2fcc": { + "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": "" + } + }, + "f2152636947d427bb30f8c8a77247dd8": { + "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 + } + }, + "6693c870f6494f40a3b64b021960d2b0": { + "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": "" + } + }, + "0921ba6b58d2433c82dac8fe3f2f4158": { + "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_16a508809c504a8591adc161bd2c35b5", + "IPY_MODEL_5c9e18ad9ecd480d876fcdf4cbf5d841", + "IPY_MODEL_51a8fd04189b4cf488369e67c82588af" + ], + "layout": "IPY_MODEL_fb5ed0aaa3c34f47bf47c67472821422" + } + }, + "16a508809c504a8591adc161bd2c35b5": { + "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_bd2ea535cd6d432daf978af539e26344", + "placeholder": "​", + "style": "IPY_MODEL_2478bca9c91148268d1df4d27b0cdd72", + "value": "Loading weights: 100%" + } + }, + "5c9e18ad9ecd480d876fcdf4cbf5d841": { + "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_6658e6702c974a6fa136992aa7fd0c90", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_9d3470ae0a0b4c93bcf93b604d70e64f", + "value": 25 + } + }, + "51a8fd04189b4cf488369e67c82588af": { + "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_74e20f0f820b42f5875099ec103c6b6e", + "placeholder": "​", + "style": "IPY_MODEL_f0c276ad409f4cac9a52cc38825ba7da", + "value": " 25/25 [00:00<00:00, 2673.51it/s, Materializing param=pooler.weight]" + } + }, + "fb5ed0aaa3c34f47bf47c67472821422": { + "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 + } + }, + "bd2ea535cd6d432daf978af539e26344": { + "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 + } + }, + "2478bca9c91148268d1df4d27b0cdd72": { + "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": "" + } + }, + "6658e6702c974a6fa136992aa7fd0c90": { + "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 + } + }, + "9d3470ae0a0b4c93bcf93b604d70e64f": { + "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": "" + } + }, + "74e20f0f820b42f5875099ec103c6b6e": { + "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 + } + }, + "f0c276ad409f4cac9a52cc38825ba7da": { + "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": "" + } + }, + "3d89b978a63b4d4aa2b82cce48d99cc5": { + "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_ee0532f1b4ee44c0a507878f40d0c2b9", + "IPY_MODEL_cf7988090d4e4a32b48f80b4766993cf", + "IPY_MODEL_f8a7d1fa879941eda3861daca4556620" + ], + "layout": "IPY_MODEL_21d612aac9f64fdaa29525fa13f34d07" + } + }, + "ee0532f1b4ee44c0a507878f40d0c2b9": { + "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_e473894f50c04e70a4ae7b8cd34f1ac0", + "placeholder": "​", + "style": "IPY_MODEL_c93ad11d249b445b8941b5f628575fd6", + "value": "Loading weights: 100%" + } + }, + "cf7988090d4e4a32b48f80b4766993cf": { + "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_bee5c523c01b40ea8cb919f13273cde1", + "max": 100, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e1e68445ab5c47c0ba6ee4636ad070a9", + "value": 100 + } + }, + "f8a7d1fa879941eda3861daca4556620": { + "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_228b6eef214d4925960ee8bd32ef670e", + "placeholder": "​", + "style": "IPY_MODEL_6871542b5d374ee49e7082b35891f4dd", + "value": " 100/100 [00:00<00:00, 3793.18it/s, Materializing param=transformer.layer.5.sa_layer_norm.weight]" + } + }, + "21d612aac9f64fdaa29525fa13f34d07": { + "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 + } + }, + "e473894f50c04e70a4ae7b8cd34f1ac0": { + "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 + } + }, + "c93ad11d249b445b8941b5f628575fd6": { + "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": "" + } + }, + "bee5c523c01b40ea8cb919f13273cde1": { + "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 + } + }, + "e1e68445ab5c47c0ba6ee4636ad070a9": { + "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": "" + } + }, + "228b6eef214d4925960ee8bd32ef670e": { + "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 + } + }, + "6871542b5d374ee49e7082b35891f4dd": { + "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": "" + } + }, + "8744db3b89ed47af85737050cbb8082a": { + "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_106f34fdd39c4c1eb65e2a96f3d3e18f", + "IPY_MODEL_54ae3f40c1e04f799031b3b23aa69b68", + "IPY_MODEL_11cf9cc56af94afe9b13a2bf8eafc0d9" + ], + "layout": "IPY_MODEL_e82698ea88cf43a9b1b69362d3c2496d" + } + }, + "106f34fdd39c4c1eb65e2a96f3d3e18f": { + "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_a1b5cc871f1649829fe59becd2b04792", + "placeholder": "​", + "style": "IPY_MODEL_a7f543ed5e724fd99292d7fe08b6f302", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "54ae3f40c1e04f799031b3b23aa69b68": { + "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_0c6748119cfa4f4c89e566dc4a1b8c11", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_68ac7165011f436e882e94706fb99332", + "value": 1 + } + }, + "11cf9cc56af94afe9b13a2bf8eafc0d9": { + "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_2e410f39fc264bfc8f3f2346a424eb9d", + "placeholder": "​", + "style": "IPY_MODEL_dfcbb71207824172985b3af4a2ec0e93", + "value": "  104MB /  104MB, 47.2MB/s  " + } + }, + "e82698ea88cf43a9b1b69362d3c2496d": { + "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 + } + }, + "a1b5cc871f1649829fe59becd2b04792": { + "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 + } + }, + "a7f543ed5e724fd99292d7fe08b6f302": { + "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": "" + } + }, + "0c6748119cfa4f4c89e566dc4a1b8c11": { + "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": "20px" + } + }, + "68ac7165011f436e882e94706fb99332": { + "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": "" + } + }, + "2e410f39fc264bfc8f3f2346a424eb9d": { + "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 + } + }, + "dfcbb71207824172985b3af4a2ec0e93": { + "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": "" + } + }, + "9ef68427e6664008b75c411605f0f47e": { + "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_ab339d6a735f486e90054eb1a5b2e235", + "IPY_MODEL_d94da65abaa64ca49b1689a11fa6caa6", + "IPY_MODEL_37fd8d5feff7495eb523b32da79005f8" + ], + "layout": "IPY_MODEL_ec4c6fe1342e4d009c8c1369f39a89cc" + } + }, + "ab339d6a735f486e90054eb1a5b2e235": { + "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_58c387388e6741e49c1584643640e5cb", + "placeholder": "​", + "style": "IPY_MODEL_e4e323691a2446779eb1357eb6e72476", + "value": "New Data Upload               : 100%" + } + }, + "d94da65abaa64ca49b1689a11fa6caa6": { + "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_8153340471044a9883d099762387e4ba", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6e8a6d9279704915ae1313d3ba4e09a2", + "value": 1 + } + }, + "37fd8d5feff7495eb523b32da79005f8": { + "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_9ccb55ff730f4584b3992fd03b907209", + "placeholder": "​", + "style": "IPY_MODEL_9cb6891ec0e847dca53602db6c1bf339", + "value": " 56.2MB / 56.2MB, 25.5MB/s  " + } + }, + "ec4c6fe1342e4d009c8c1369f39a89cc": { + "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 + } + }, + "58c387388e6741e49c1584643640e5cb": { + "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 + } + }, + "e4e323691a2446779eb1357eb6e72476": { + "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": "" + } + }, + "8153340471044a9883d099762387e4ba": { + "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": "20px" + } + }, + "6e8a6d9279704915ae1313d3ba4e09a2": { + "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": "" + } + }, + "9ccb55ff730f4584b3992fd03b907209": { + "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 + } + }, + "9cb6891ec0e847dca53602db6c1bf339": { + "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": "" + } + }, + "b425e5326180478ebebd37357439ef9c": { + "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_1b85ac482ac241d1a40eb48a5abce184", + "IPY_MODEL_f53ed7170069443da8e5eb0719795c27", + "IPY_MODEL_5e71fb0732624391a00b544a83283a64" + ], + "layout": "IPY_MODEL_f9e1f372feb2499e89ac0ea92221db98" + } + }, + "1b85ac482ac241d1a40eb48a5abce184": { + "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_3cd2df13b29942d5920f41565a2e4d13", + "placeholder": "​", + "style": "IPY_MODEL_979b2efed6af4d89a4f9cc4f83f977dd", + "value": "  /tmp/model.safetensors      : 100%" + } + }, + "f53ed7170069443da8e5eb0719795c27": { + "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_24ae41ab73834091900e4c9a6beaaf57", + "max": 103840272, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_a8b4fa33667b4dbbbf61216a3669f6fe", + "value": 103840272 + } + }, + "5e71fb0732624391a00b544a83283a64": { + "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_53b21868f8124bc2ac5c09f7ba0bad8e", + "placeholder": "​", + "style": "IPY_MODEL_224b422deef94653a06844e03ffea646", + "value": "  104MB /  104MB            " + } + }, + "f9e1f372feb2499e89ac0ea92221db98": { + "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 + } + }, + "3cd2df13b29942d5920f41565a2e4d13": { + "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 + } + }, + "979b2efed6af4d89a4f9cc4f83f977dd": { + "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": "" + } + }, + "24ae41ab73834091900e4c9a6beaaf57": { + "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 + } + }, + "a8b4fa33667b4dbbbf61216a3669f6fe": { + "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": "" + } + }, + "53b21868f8124bc2ac5c09f7ba0bad8e": { + "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 + } + }, + "224b422deef94653a06844e03ffea646": { + "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": "" + } + }, + "87830f32e51440729bcf1bc9c6eac889": { + "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_f71d86f36a1641239b36135fb8362989", + "IPY_MODEL_b47124c24ce8482eacf647b5089a6481", + "IPY_MODEL_08a1dc9aec854459b5ec00881c5f0e0c" + ], + "layout": "IPY_MODEL_337fd2e144c441898e163e2009b4e456" + } + }, + "f71d86f36a1641239b36135fb8362989": { + "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_346119a3cb504f4fab5a4336c206d97c", + "placeholder": "​", + "style": "IPY_MODEL_a90ca41cb45449fca8a2fa7d5bb55dbb", + "value": "Loading weights: 100%" + } + }, + "b47124c24ce8482eacf647b5089a6481": { + "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_1d9d88d9db084693abcce10e8ebc7beb", + "max": 82, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_317cfe6702af47ff8831b79c74e6b931", + "value": 82 + } + }, + "08a1dc9aec854459b5ec00881c5f0e0c": { + "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_0d12368b5c104c13a133bd0845937111", + "placeholder": "​", + "style": "IPY_MODEL_84d6a8ba7b9b4eaaa96a343d8477648c", + "value": " 82/82 [00:00<00:00, 812.15it/s, Materializing param=token_emb.weight]" + } + }, + "337fd2e144c441898e163e2009b4e456": { + "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 + } + }, + "346119a3cb504f4fab5a4336c206d97c": { + "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 + } + }, + "a90ca41cb45449fca8a2fa7d5bb55dbb": { + "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": "" + } + }, + "1d9d88d9db084693abcce10e8ebc7beb": { + "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 + } + }, + "317cfe6702af47ff8831b79c74e6b931": { + "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": "" + } + }, + "0d12368b5c104c13a133bd0845937111": { + "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 + } + }, + "84d6a8ba7b9b4eaaa96a343d8477648c": { + "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": "" + } + }, + "fc4154afb1d54ade92bd8f1f03d85d22": { + "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_242d8f5bd00b4a8a82d13cab2fedbb11", + "IPY_MODEL_5d6c91de29bd4689b951e8f8da31723a", + "IPY_MODEL_aadae893f97046449cfdb92fe54619f1" + ], + "layout": "IPY_MODEL_dd2b7d9da2044054b13463ef08789981" + } + }, + "242d8f5bd00b4a8a82d13cab2fedbb11": { + "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_e3ac055fbe0a476ba92c9ac428c46d3c", + "placeholder": "​", + "style": "IPY_MODEL_56dc19f3b096416a84a308c5ab866578", + "value": "Loading weights: 100%" + } + }, + "5d6c91de29bd4689b951e8f8da31723a": { + "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_e2e5e2362d904f108884756cda1d6518", + "max": 82, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0d655e4f97c24655a08001461a8824c2", + "value": 82 + } + }, + "aadae893f97046449cfdb92fe54619f1": { + "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_1019f3ce8439413b975e1ff6444cf61a", + "placeholder": "​", + "style": "IPY_MODEL_e4b69e37104d4a60877e9d29062baea1", + "value": " 82/82 [00:00<00:00, 703.18it/s, Materializing param=token_emb.weight]" + } + }, + "dd2b7d9da2044054b13463ef08789981": { + "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 + } + }, + "e3ac055fbe0a476ba92c9ac428c46d3c": { + "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 + } + }, + "56dc19f3b096416a84a308c5ab866578": { + "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": "" + } + }, + "e2e5e2362d904f108884756cda1d6518": { + "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 + } + }, + "0d655e4f97c24655a08001461a8824c2": { + "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": "" + } + }, + "1019f3ce8439413b975e1ff6444cf61a": { + "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 + } + }, + "e4b69e37104d4a60877e9d29062baea1": { + "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": "" + } + }, + "e1a8d24aa52a43ce816a516d858fdf66": { + "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_55ced514f9894c65bcfb78f8f64b72b9", + "IPY_MODEL_961c1273295040a98df4bcb424d83ef0", + "IPY_MODEL_082caf846d1e4ffa8f10e99d4d1ef285" + ], + "layout": "IPY_MODEL_e7365eb9ad9449e8a34d9ebcaa5590f4" + } + }, + "55ced514f9894c65bcfb78f8f64b72b9": { + "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_3ca334c988de406fab7c123a28188fba", + "placeholder": "​", + "style": "IPY_MODEL_f1585bd61f234038a0482ce2a64815c1", + "value": "Loading weights: 100%" + } + }, + "961c1273295040a98df4bcb424d83ef0": { + "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_2aea60dac24e4bb7a11a3d95e1420a93", + "max": 82, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b8190678a51f4e41ab1a9b9625c3a9e6", + "value": 82 + } + }, + "082caf846d1e4ffa8f10e99d4d1ef285": { + "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_bd0484aa4a96450d8d820bddf38bb36e", + "placeholder": "​", + "style": "IPY_MODEL_f517d583ed7046dfb6fa9c03d355d034", + "value": " 82/82 [00:00<00:00, 3663.38it/s, Materializing param=token_emb.weight]" + } + }, + "e7365eb9ad9449e8a34d9ebcaa5590f4": { + "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 + } + }, + "3ca334c988de406fab7c123a28188fba": { + "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 + } + }, + "f1585bd61f234038a0482ce2a64815c1": { + "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": "" + } + }, + "2aea60dac24e4bb7a11a3d95e1420a93": { + "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 + } + }, + "b8190678a51f4e41ab1a9b9625c3a9e6": { + "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": "" + } + }, + "bd0484aa4a96450d8d820bddf38bb36e": { + "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 + } + }, + "f517d583ed7046dfb6fa9c03d355d034": { + "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": "" + } + }, + "9ce5f076aa2c4cab9375fb3d7a4289b1": { + "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_7b6c598e02e7484b90b0c7178a82a2fc", + "IPY_MODEL_4ad5d6904a0348fea992da43585b00ef", + "IPY_MODEL_249ff1a4e6b94a9185642e20c5f1bda8" + ], + "layout": "IPY_MODEL_b13bbe61c7804367b740463d6f24ac3b" + } + }, + "7b6c598e02e7484b90b0c7178a82a2fc": { + "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_413f5527831c49829a96e0ac303d4a1a", + "placeholder": "​", + "style": "IPY_MODEL_fb1f6e367eec4cdbb1be205bb76b7b83", + "value": "Loading weights: 100%" + } + }, + "4ad5d6904a0348fea992da43585b00ef": { + "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_845d21a05b404a1aae834514a3bdddfa", + "max": 82, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ecc945a6b0e344548301f6f2b3477844", + "value": 82 + } + }, + "249ff1a4e6b94a9185642e20c5f1bda8": { + "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_37e0ad2f47074b4795a717a64a22a254", + "placeholder": "​", + "style": "IPY_MODEL_2018eec3287f46a7a2ae822f560e6f96", + "value": " 82/82 [00:00<00:00, 3603.35it/s, Materializing param=token_emb.weight]" + } + }, + "b13bbe61c7804367b740463d6f24ac3b": { + "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 + } + }, + "413f5527831c49829a96e0ac303d4a1a": { + "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 + } + }, + "fb1f6e367eec4cdbb1be205bb76b7b83": { + "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": "" + } + }, + "845d21a05b404a1aae834514a3bdddfa": { + "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 + } + }, + "ecc945a6b0e344548301f6f2b3477844": { + "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": "" + } + }, + "37e0ad2f47074b4795a717a64a22a254": { + "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 + } + }, + "2018eec3287f46a7a2ae822f560e6f96": { + "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": "" + } + }, + "b7f0462796534c0c887869dd4ee536b6": { + "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_b04346b7cdd44d1ca83bb3f9164d34b7", + "IPY_MODEL_4c0fd3ff39ec45029ba60e6cef15cd2b", + "IPY_MODEL_08611fc88d5143e6b79234aaae1886ef" + ], + "layout": "IPY_MODEL_5420ef57acbd4189832bce4f46fb3c78" + } + }, + "b04346b7cdd44d1ca83bb3f9164d34b7": { + "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_e6d9053451594834970c6a48833acbe0", + "placeholder": "​", + "style": "IPY_MODEL_0917a8224cb645aeaa9beefd8dddcdda", + "value": "Loading weights: 100%" + } + }, + "4c0fd3ff39ec45029ba60e6cef15cd2b": { + "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_9027d19005a5449189ec7a1440f3d9b7", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_798c1ea91e06483389133dc069842676", + "value": 199 + } + }, + "08611fc88d5143e6b79234aaae1886ef": { + "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_66168ca2364347aab0174721ab52e400", + "placeholder": "​", + "style": "IPY_MODEL_621c518987f74847a99d25597315e17c", + "value": " 199/199 [00:00<00:00, 4241.54it/s, Materializing param=pooler.dense.weight]" + } + }, + "5420ef57acbd4189832bce4f46fb3c78": { + "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 + } + }, + "e6d9053451594834970c6a48833acbe0": { + "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 + } + }, + "0917a8224cb645aeaa9beefd8dddcdda": { + "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": "" + } + }, + "9027d19005a5449189ec7a1440f3d9b7": { + "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 + } + }, + "798c1ea91e06483389133dc069842676": { + "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": "" + } + }, + "66168ca2364347aab0174721ab52e400": { + "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 + } + }, + "621c518987f74847a99d25597315e17c": { + "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": "" + } + }, + "0074e789f07c4864a2d0bdae43a0a950": { + "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_fa510b450a2e4b8d929cf23ba0ce8e6f", + "IPY_MODEL_352f5aa1999143e1840fd185b76a1980", + "IPY_MODEL_4884f934d3be4876b99112e5d9cc42ba" + ], + "layout": "IPY_MODEL_e45c8496bebd491ca4343026f860f6f6" + } + }, + "fa510b450a2e4b8d929cf23ba0ce8e6f": { + "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_1560366419274f78882efd9df291137b", + "placeholder": "​", + "style": "IPY_MODEL_262abd0b7042430ab85f64c184f21fc5", + "value": "Loading weights: 100%" + } + }, + "352f5aa1999143e1840fd185b76a1980": { + "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_7235412109f54367a9d19b29b8481df6", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_062b0957ba7e49bb9fb449da603061ee", + "value": 134 + } + }, + "4884f934d3be4876b99112e5d9cc42ba": { + "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_0638d340162c4f0aac6d78f0d7856d59", + "placeholder": "​", + "style": "IPY_MODEL_28cc211bf2194de3a931c0f637d8edc9", + "value": " 134/134 [00:00<00:00, 4034.35it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "e45c8496bebd491ca4343026f860f6f6": { + "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 + } + }, + "1560366419274f78882efd9df291137b": { + "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 + } + }, + "262abd0b7042430ab85f64c184f21fc5": { + "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": "" + } + }, + "7235412109f54367a9d19b29b8481df6": { + "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 + } + }, + "062b0957ba7e49bb9fb449da603061ee": { + "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": "" + } + }, + "0638d340162c4f0aac6d78f0d7856d59": { + "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 + } + }, + "28cc211bf2194de3a931c0f637d8edc9": { + "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": "" + } + }, + "fb248fb346564862b9c633d195d88d2a": { + "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_68cc85aa9433422d925937b245bad649", + "IPY_MODEL_ed527cdca7de4f569ae2e879ae2c5cc4", + "IPY_MODEL_3926e9cb649844cbaed16d16af96490d" + ], + "layout": "IPY_MODEL_19dff49386d24a4cb9940d1c41b6d87c" + } + }, + "68cc85aa9433422d925937b245bad649": { + "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_75912febbe024523ab196ebee515abe3", + "placeholder": "​", + "style": "IPY_MODEL_a492e1fac92342dab64eac36dd2694ea", + "value": "Loading weights: 100%" + } + }, + "ed527cdca7de4f569ae2e879ae2c5cc4": { + "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_e00e46d477814ff3a15d3a0549cadcb4", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_2665800f69444ac5b9f7ba2d70eab01d", + "value": 199 + } + }, + "3926e9cb649844cbaed16d16af96490d": { + "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_93257a415826429f9de685655370237e", + "placeholder": "​", + "style": "IPY_MODEL_c01652b8718e4a80b8c92c82a1fc9f49", + "value": " 199/199 [00:00<00:00, 4355.73it/s, Materializing param=pooler.dense.weight]" + } + }, + "19dff49386d24a4cb9940d1c41b6d87c": { + "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 + } + }, + "75912febbe024523ab196ebee515abe3": { + "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 + } + }, + "a492e1fac92342dab64eac36dd2694ea": { + "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": "" + } + }, + "e00e46d477814ff3a15d3a0549cadcb4": { + "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 + } + }, + "2665800f69444ac5b9f7ba2d70eab01d": { + "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": "" + } + }, + "93257a415826429f9de685655370237e": { + "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 + } + }, + "c01652b8718e4a80b8c92c82a1fc9f49": { + "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": "" + } + }, + "5c2fe3ce505c425b8a5133636fa3e59a": { + "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_8eaf6eedc46144ea828b8cf1bf3c177a", + "IPY_MODEL_9b4aa70316fd4b30acc0a957b2e024dd", + "IPY_MODEL_dd4163db08b94b88bbdb6f2e914d91a4" + ], + "layout": "IPY_MODEL_17938cc8e1184b3ca9db7c72fe5c40fe" + } + }, + "8eaf6eedc46144ea828b8cf1bf3c177a": { + "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_9549dcb9f1604714a906e581101174fc", + "placeholder": "​", + "style": "IPY_MODEL_493fc58cc1084e00b4aef9e1a4ce7687", + "value": "Loading weights: 100%" + } + }, + "9b4aa70316fd4b30acc0a957b2e024dd": { + "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_73fdc6a1594c47de86c5124e192bc43f", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6fbf05deedf94711a079726a73811f52", + "value": 134 + } + }, + "dd4163db08b94b88bbdb6f2e914d91a4": { + "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_3f36a310c43c4012a2a0c19294fe0e13", + "placeholder": "​", + "style": "IPY_MODEL_11b3b34f9ffc4e879152200104225df1", + "value": " 134/134 [00:00<00:00, 4050.54it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "17938cc8e1184b3ca9db7c72fe5c40fe": { + "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 + } + }, + "9549dcb9f1604714a906e581101174fc": { + "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 + } + }, + "493fc58cc1084e00b4aef9e1a4ce7687": { + "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": "" + } + }, + "73fdc6a1594c47de86c5124e192bc43f": { + "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 + } + }, + "6fbf05deedf94711a079726a73811f52": { + "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": "" + } + }, + "3f36a310c43c4012a2a0c19294fe0e13": { + "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 + } + }, + "11b3b34f9ffc4e879152200104225df1": { + "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": "" + } + }, + "96e6b188456a440491abf1b9f97e0f1b": { + "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_2bbf292baba24d83bc6969f17945ebdf", + "IPY_MODEL_bfbcf1c77d2e4c6dab61ecdb8b32f45b", + "IPY_MODEL_c9380e44797d4631bacdce608af8501d" + ], + "layout": "IPY_MODEL_e4d3027a8ed1412693b1e8dd601c06ab" + } + }, + "2bbf292baba24d83bc6969f17945ebdf": { + "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_f63ba9edd18d483795ebaa2ac160dbed", + "placeholder": "​", + "style": "IPY_MODEL_2d01e0821f704e09945bcd17419c11e2", + "value": "Loading weights: 100%" + } + }, + "bfbcf1c77d2e4c6dab61ecdb8b32f45b": { + "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_e89e69591d21429fa8007b0fa576d61c", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5cdba3bf3b5f4d259ebd63ecab967fc2", + "value": 199 + } + }, + "c9380e44797d4631bacdce608af8501d": { + "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_169c779b68ca4da6a16eb88b87e0cf62", + "placeholder": "​", + "style": "IPY_MODEL_c4bdcb3665724031a1e7f1f06cfd496e", + "value": " 199/199 [00:00<00:00, 4225.03it/s, Materializing param=pooler.dense.weight]" + } + }, + "e4d3027a8ed1412693b1e8dd601c06ab": { + "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 + } + }, + "f63ba9edd18d483795ebaa2ac160dbed": { + "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 + } + }, + "2d01e0821f704e09945bcd17419c11e2": { + "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": "" + } + }, + "e89e69591d21429fa8007b0fa576d61c": { + "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 + } + }, + "5cdba3bf3b5f4d259ebd63ecab967fc2": { + "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": "" + } + }, + "169c779b68ca4da6a16eb88b87e0cf62": { + "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 + } + }, + "c4bdcb3665724031a1e7f1f06cfd496e": { + "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": "" + } + }, + "1be48dcc1d5a44af98283079bf2e251c": { + "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_d00dd42d3bd5429d86f33edaa9a85b8b", + "IPY_MODEL_c219a9706b674b7aa89c4d5bc11e6df3", + "IPY_MODEL_1f772e7517fe4ef5bbcc657a384d37e8" + ], + "layout": "IPY_MODEL_eadbb240982c4fb6aa19bf61ae99c672" + } + }, + "d00dd42d3bd5429d86f33edaa9a85b8b": { + "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_b06994fee46d41d7930ce03d439a1ae9", + "placeholder": "​", + "style": "IPY_MODEL_0c37b804865244e9b4db2aac3e277a6e", + "value": "Loading weights: 100%" + } + }, + "c219a9706b674b7aa89c4d5bc11e6df3": { + "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_3284c3d7ea4342dbae5387da632d73f4", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_3bbc747f5ccf4acab6ae9c38506fef28", + "value": 134 + } + }, + "1f772e7517fe4ef5bbcc657a384d37e8": { + "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_4af6797d8e78453982f35694c60785ef", + "placeholder": "​", + "style": "IPY_MODEL_a1f0417b8f464996b1c93dc219c3d22f", + "value": " 134/134 [00:00<00:00, 3994.92it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "eadbb240982c4fb6aa19bf61ae99c672": { + "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 + } + }, + "b06994fee46d41d7930ce03d439a1ae9": { + "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 + } + }, + "0c37b804865244e9b4db2aac3e277a6e": { + "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": "" + } + }, + "3284c3d7ea4342dbae5387da632d73f4": { + "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 + } + }, + "3bbc747f5ccf4acab6ae9c38506fef28": { + "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": "" + } + }, + "4af6797d8e78453982f35694c60785ef": { + "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 + } + }, + "a1f0417b8f464996b1c93dc219c3d22f": { + "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": "" + } + }, + "3fc59d54447c48cf95c06fa0b7c44f81": { + "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_97a9f9336df3492187221ef28aa2be6c", + "IPY_MODEL_44a2bc1c6841464ab8a914aa7907f0e9", + "IPY_MODEL_2b914a27903c485faa75b8f755bffe6e" + ], + "layout": "IPY_MODEL_4caeb796ecc642148f1921227a0ffa76" + } + }, + "97a9f9336df3492187221ef28aa2be6c": { + "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_e892f44899514757af1188a6e80b7382", + "placeholder": "​", + "style": "IPY_MODEL_5b31210bff804a90802064e36ed51938", + "value": "Loading weights: 100%" + } + }, + "44a2bc1c6841464ab8a914aa7907f0e9": { + "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_a46ef317e76345dda61ce0ce5eb61181", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_650619cadc124853993e836b3fae340e", + "value": 199 + } + }, + "2b914a27903c485faa75b8f755bffe6e": { + "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_df54dc7f2cfb496d923f85aa9aa719f7", + "placeholder": "​", + "style": "IPY_MODEL_9258ee70426b4117ae3ae148f6332581", + "value": " 199/199 [00:00<00:00, 4196.37it/s, Materializing param=pooler.dense.weight]" + } + }, + "4caeb796ecc642148f1921227a0ffa76": { + "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 + } + }, + "e892f44899514757af1188a6e80b7382": { + "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 + } + }, + "5b31210bff804a90802064e36ed51938": { + "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": "" + } + }, + "a46ef317e76345dda61ce0ce5eb61181": { + "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 + } + }, + "650619cadc124853993e836b3fae340e": { + "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": "" + } + }, + "df54dc7f2cfb496d923f85aa9aa719f7": { + "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 + } + }, + "9258ee70426b4117ae3ae148f6332581": { + "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": "" + } + }, + "9faa8b9c10da45b6b3088d2246dab05b": { + "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_926ffa40ae224a61927901a0b7524251", + "IPY_MODEL_27802385da91475ab1b765fee5519d87", + "IPY_MODEL_7265f91c04194da3a4abf7c7b8bedff2" + ], + "layout": "IPY_MODEL_c2c5c0a854ee477680bdffb98b0d8a89" + } + }, + "926ffa40ae224a61927901a0b7524251": { + "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_75a32209b14e466ab66498a2a7894b28", + "placeholder": "​", + "style": "IPY_MODEL_ae3a834c74d34f7e842764df0ea2ae44", + "value": "Loading weights: 100%" + } + }, + "27802385da91475ab1b765fee5519d87": { + "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_41a228039e1843dc92e771715040d525", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b2c514fc44ba465bb6d77c6097f3ddf7", + "value": 134 + } + }, + "7265f91c04194da3a4abf7c7b8bedff2": { + "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_760d3b2292ba4f67a9927387024169d7", + "placeholder": "​", + "style": "IPY_MODEL_7da5505ebc93451a81f4c5b7b8a72ee1", + "value": " 134/134 [00:00<00:00, 3900.16it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "c2c5c0a854ee477680bdffb98b0d8a89": { + "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 + } + }, + "75a32209b14e466ab66498a2a7894b28": { + "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 + } + }, + "ae3a834c74d34f7e842764df0ea2ae44": { + "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": "" + } + }, + "41a228039e1843dc92e771715040d525": { + "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 + } + }, + "b2c514fc44ba465bb6d77c6097f3ddf7": { + "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": "" + } + }, + "760d3b2292ba4f67a9927387024169d7": { + "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 + } + }, + "7da5505ebc93451a81f4c5b7b8a72ee1": { + "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": "" + } + }, + "c3c92d93759d4f1bacf7b0c9e82fea8c": { + "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_d81ed69dac29481098abe7e3367dd946", + "IPY_MODEL_eae680c132324913a1bd220bc99343b1", + "IPY_MODEL_a3ae20ce4d5d4a8cb5aaa2792f4ac8fb" + ], + "layout": "IPY_MODEL_d969fd9386254c979c9195b7ba4fc9da" + } + }, + "d81ed69dac29481098abe7e3367dd946": { + "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_556728e1ff704dcd8b8110d445984172", + "placeholder": "​", + "style": "IPY_MODEL_7f0c93c28b2c4b529e15a5baa045c9f2", + "value": "Loading weights: 100%" + } + }, + "eae680c132324913a1bd220bc99343b1": { + "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_7c5142b27e8042a1861a87fb880a3ffd", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_1f1255df608f4d3eaec9a2d618a73c2e", + "value": 199 + } + }, + "a3ae20ce4d5d4a8cb5aaa2792f4ac8fb": { + "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_d4cc4e02fcf4457d9c7641457c33851e", + "placeholder": "​", + "style": "IPY_MODEL_9ba7c36a314741cf837f9fa47e3c4d4e", + "value": " 199/199 [00:00<00:00, 4038.86it/s, Materializing param=pooler.dense.weight]" + } + }, + "d969fd9386254c979c9195b7ba4fc9da": { + "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 + } + }, + "556728e1ff704dcd8b8110d445984172": { + "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 + } + }, + "7f0c93c28b2c4b529e15a5baa045c9f2": { + "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": "" + } + }, + "7c5142b27e8042a1861a87fb880a3ffd": { + "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 + } + }, + "1f1255df608f4d3eaec9a2d618a73c2e": { + "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": "" + } + }, + "d4cc4e02fcf4457d9c7641457c33851e": { + "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 + } + }, + "9ba7c36a314741cf837f9fa47e3c4d4e": { + "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": "" + } + }, + "9ee9cc7e622641d19fc3d3933f01f69d": { + "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_c74ba5f763d94eb8803dfbf92ffe42b7", + "IPY_MODEL_ba31d5f5e2614febb1d312c240f7ff3a", + "IPY_MODEL_c0391a88ed0940b48f567e11db716d61" + ], + "layout": "IPY_MODEL_937863ac9ff4471797515158207b6ef6" + } + }, + "c74ba5f763d94eb8803dfbf92ffe42b7": { + "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_e3dcb07e408d441f926c7778dc0a3bba", + "placeholder": "​", + "style": "IPY_MODEL_b11dbe41f79e4e178ef2e5bc75271414", + "value": "Loading weights: 100%" + } + }, + "ba31d5f5e2614febb1d312c240f7ff3a": { + "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_43c9d5efd89a4a5ea4ffeb70c55ed556", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_a15366153a1a4dd1a034173435d3099e", + "value": 134 + } + }, + "c0391a88ed0940b48f567e11db716d61": { + "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_23600aa3bd934c5abe36d4ea031ae839", + "placeholder": "​", + "style": "IPY_MODEL_51e4afb3f3cc4698ada6e1ae1de6bae9", + "value": " 134/134 [00:00<00:00, 4016.84it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "937863ac9ff4471797515158207b6ef6": { + "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 + } + }, + "e3dcb07e408d441f926c7778dc0a3bba": { + "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 + } + }, + "b11dbe41f79e4e178ef2e5bc75271414": { + "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": "" + } + }, + "43c9d5efd89a4a5ea4ffeb70c55ed556": { + "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 + } + }, + "a15366153a1a4dd1a034173435d3099e": { + "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": "" + } + }, + "23600aa3bd934c5abe36d4ea031ae839": { + "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 + } + }, + "51e4afb3f3cc4698ada6e1ae1de6bae9": { + "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": "" + } + }, + "adf5073645bd4927b23e9e2f245e57bc": { + "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_41fa9a998197421db2e37852bd643d20", + "IPY_MODEL_3c28a35ed90345d5983b6e7910e0076e", + "IPY_MODEL_19460177f35546d7af56905bfa79dbe9" + ], + "layout": "IPY_MODEL_609d8ace2dca4f9aa239ece209512e3c" + } + }, + "41fa9a998197421db2e37852bd643d20": { + "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_4d1db66017314a47917b4f7e370b2da3", + "placeholder": "​", + "style": "IPY_MODEL_78512c07da324a4d91baff786f68f79b", + "value": "Loading weights: 100%" + } + }, + "3c28a35ed90345d5983b6e7910e0076e": { + "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_e6e0a63481964fbe982e006f2d855bbb", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ebcea318e84145b6a5d238df793755da", + "value": 199 + } + }, + "19460177f35546d7af56905bfa79dbe9": { + "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_a706ddc178fa405a9a058283431ae4a7", + "placeholder": "​", + "style": "IPY_MODEL_536780f5538e4c148d6c08014a76f670", + "value": " 199/199 [00:00<00:00, 4473.60it/s, Materializing param=pooler.dense.weight]" + } + }, + "609d8ace2dca4f9aa239ece209512e3c": { + "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 + } + }, + "4d1db66017314a47917b4f7e370b2da3": { + "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 + } + }, + "78512c07da324a4d91baff786f68f79b": { + "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": "" + } + }, + "e6e0a63481964fbe982e006f2d855bbb": { + "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 + } + }, + "ebcea318e84145b6a5d238df793755da": { + "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": "" + } + }, + "a706ddc178fa405a9a058283431ae4a7": { + "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 + } + }, + "536780f5538e4c148d6c08014a76f670": { + "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": "" + } + }, + "8acc67491474462e9e062baca52c04a2": { + "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_2eb6770aee654f388176238f09778b30", + "IPY_MODEL_95b5c43c48594913ae0d8e315a9c80f9", + "IPY_MODEL_fb6a8aa0a91b4bb0adc448ad84b1428b" + ], + "layout": "IPY_MODEL_2e96ea3acbad4fee9d528107a1e527af" + } + }, + "2eb6770aee654f388176238f09778b30": { + "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_aa01aa0a79e1496c82c8a0278cf9c3da", + "placeholder": "​", + "style": "IPY_MODEL_3440899e994a4b68971ad5924d9db1ad", + "value": "Loading weights: 100%" + } + }, + "95b5c43c48594913ae0d8e315a9c80f9": { + "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_e761b132c88240549199f12afaf5a650", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_cb3b9be5979a4f9cb8a8186d1e324cdc", + "value": 134 + } + }, + "fb6a8aa0a91b4bb0adc448ad84b1428b": { + "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_b343f92042b94bc78eb0af35fd65a1bf", + "placeholder": "​", + "style": "IPY_MODEL_4d73be6b0016448da67fa630570807f2", + "value": " 134/134 [00:00<00:00, 4002.54it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "2e96ea3acbad4fee9d528107a1e527af": { + "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 + } + }, + "aa01aa0a79e1496c82c8a0278cf9c3da": { + "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 + } + }, + "3440899e994a4b68971ad5924d9db1ad": { + "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": "" + } + }, + "e761b132c88240549199f12afaf5a650": { + "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 + } + }, + "cb3b9be5979a4f9cb8a8186d1e324cdc": { + "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": "" + } + }, + "b343f92042b94bc78eb0af35fd65a1bf": { + "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 + } + }, + "4d73be6b0016448da67fa630570807f2": { + "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": "" + } + }, + "b63e64cfb73141dca1082ac02358b288": { + "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_fa1545986ad94525a89ee64cd3ed2341", + "IPY_MODEL_62da4d748f644d36a45c1a9306c226f2", + "IPY_MODEL_1da92130d7114b55a8a4fc691337691e" + ], + "layout": "IPY_MODEL_3e8ee8649aef4331af229d584a6bccba" + } + }, + "fa1545986ad94525a89ee64cd3ed2341": { + "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_1b06a9738e444782b7885db78dfd30c9", + "placeholder": "​", + "style": "IPY_MODEL_1114ece9b9884973a790259962983edb", + "value": "Loading weights: 100%" + } + }, + "62da4d748f644d36a45c1a9306c226f2": { + "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_ae2c80f5d3db4b5f8b9748a34ea27906", + "max": 199, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e3c8e44dcd0e4b3b84a5c60493e8684e", + "value": 199 + } + }, + "1da92130d7114b55a8a4fc691337691e": { + "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_0fa8053b5f6340a49ed293ca538c41e0", + "placeholder": "​", + "style": "IPY_MODEL_00dfa9645b234496be78e8f8e58c8e38", + "value": " 199/199 [00:00<00:00, 4325.21it/s, Materializing param=pooler.dense.weight]" + } + }, + "3e8ee8649aef4331af229d584a6bccba": { + "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 + } + }, + "1b06a9738e444782b7885db78dfd30c9": { + "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 + } + }, + "1114ece9b9884973a790259962983edb": { + "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": "" + } + }, + "ae2c80f5d3db4b5f8b9748a34ea27906": { + "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 + } + }, + "e3c8e44dcd0e4b3b84a5c60493e8684e": { + "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": "" + } + }, + "0fa8053b5f6340a49ed293ca538c41e0": { + "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 + } + }, + "00dfa9645b234496be78e8f8e58c8e38": { + "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": "" + } + }, + "2dcc8c0441bb4966a6c9ec3b810c06fe": { + "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_e83df993ae8c4e26a9aef7c2ecafc810", + "IPY_MODEL_370d7642e679416690f7380ae3ba03b4", + "IPY_MODEL_2ece17832e524e80a1740571ea0557f8" + ], + "layout": "IPY_MODEL_1ab761ee76294cb1a049a3c672bf47af" + } + }, + "e83df993ae8c4e26a9aef7c2ecafc810": { + "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_da9b05cbed414810bab33c39f192f7e8", + "placeholder": "​", + "style": "IPY_MODEL_ed090236ac4a42598beb85e2916abe1e", + "value": "Loading weights: 100%" + } + }, + "370d7642e679416690f7380ae3ba03b4": { + "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_85c3b5826c104280bb2a3483f87bcae3", + "max": 134, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5c0e201b8dfb4db4a587d1dc0046089f", + "value": 134 + } + }, + "2ece17832e524e80a1740571ea0557f8": { + "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_ca439a1c8d9a42f1b5a3cce4e936b60d", + "placeholder": "​", + "style": "IPY_MODEL_b7fea8a491a84fb99d7e1aea9fc7f61f", + "value": " 134/134 [00:00<00:00, 3173.52it/s, Materializing param=layers.21.mlp_norm.weight]" + } + }, + "1ab761ee76294cb1a049a3c672bf47af": { + "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 + } + }, + "da9b05cbed414810bab33c39f192f7e8": { + "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 + } + }, + "ed090236ac4a42598beb85e2916abe1e": { + "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": "" + } + }, + "85c3b5826c104280bb2a3483f87bcae3": { + "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 + } + }, + "5c0e201b8dfb4db4a587d1dc0046089f": { + "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": "" + } + }, + "ca439a1c8d9a42f1b5a3cce4e936b60d": { + "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 + } + }, + "b7fea8a491a84fb99d7e1aea9fc7f61f": { + "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": "" + } + }, + "c73d64f12b3547fc8cd59b4598812465": { + "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_602a23aaf0a4401aa5e3c9263f6026c5", + "IPY_MODEL_170719a38be14c138a17a4a9c315104d", + "IPY_MODEL_3e7a63e286344af1ba9a49e7ca55380c" + ], + "layout": "IPY_MODEL_a5bb395b6b5c4c6b9008ae7efbe5f879" + } + }, + "602a23aaf0a4401aa5e3c9263f6026c5": { + "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_b00700760d5744a3abcecda5c56a6a37", + "placeholder": "​", + "style": "IPY_MODEL_7e2d5b90cb674b0aade480ca52539ab9", + "value": "config.json: 100%" + } + }, + "170719a38be14c138a17a4a9c315104d": { + "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_06d1a27817564802bd85b5144a8f66b1", + "max": 481, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c4885af32a6f45e682a15bd7b2681404", + "value": 481 + } + }, + "3e7a63e286344af1ba9a49e7ca55380c": { + "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_88a1cc0f868041bf97319dc4eb14965c", + "placeholder": "​", + "style": "IPY_MODEL_f78341ea0b984611ab0320ada1474f00", + "value": " 481/481 [00:00<00:00, 146kB/s]" + } + }, + "a5bb395b6b5c4c6b9008ae7efbe5f879": { + "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 + } + }, + "b00700760d5744a3abcecda5c56a6a37": { + "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 + } + }, + "7e2d5b90cb674b0aade480ca52539ab9": { + "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": "" + } + }, + "06d1a27817564802bd85b5144a8f66b1": { + "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 + } + }, + "c4885af32a6f45e682a15bd7b2681404": { + "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": "" + } + }, + "88a1cc0f868041bf97319dc4eb14965c": { + "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 + } + }, + "f78341ea0b984611ab0320ada1474f00": { + "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": "" + } + }, + "3d49c72bf02f46d99659693617d02bad": { + "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_a41e39e89be040b79b23ffa3c90fb41f", + "IPY_MODEL_c2d339a959a94b3b893172e6b79a46f4", + "IPY_MODEL_9a5b2d6e52b34c5e827f271d159ae4d1" + ], + "layout": "IPY_MODEL_5a47fce188ba4a54835f9c203b3161e5" + } + }, + "a41e39e89be040b79b23ffa3c90fb41f": { + "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_9d97322824c4451baba172989e0625d1", + "placeholder": "​", + "style": "IPY_MODEL_d6fdf04a597045fb80130b54ac706d95", + "value": "model.safetensors: 100%" + } + }, + "c2d339a959a94b3b893172e6b79a46f4": { + "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_78ee2e9a06c84bacb21aba15ffc4c8f4", + "max": 498818054, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_27269f63df6a4e9aa0093279f7b42d08", + "value": 498818054 + } + }, + "9a5b2d6e52b34c5e827f271d159ae4d1": { + "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_35ad86ea0710430fad41d1dc09c7855e", + "placeholder": "​", + "style": "IPY_MODEL_48db0905592b4b2abc3ab5360b00f058", + "value": " 499M/499M [00:01<00:00, 557MB/s]" + } + }, + "5a47fce188ba4a54835f9c203b3161e5": { + "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 + } + }, + "9d97322824c4451baba172989e0625d1": { + "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 + } + }, + "d6fdf04a597045fb80130b54ac706d95": { + "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": "" + } + }, + "78ee2e9a06c84bacb21aba15ffc4c8f4": { + "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 + } + }, + "27269f63df6a4e9aa0093279f7b42d08": { + "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": "" + } + }, + "35ad86ea0710430fad41d1dc09c7855e": { + "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 + } + }, + "48db0905592b4b2abc3ab5360b00f058": { + "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": "" + } + }, + "7f76e88c6b6b4df294c9e06abff02300": { + "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_fc561d3e7239434f82e5f97834882373", + "IPY_MODEL_39c70d8edbca4f71b171f8d278d3e7da", + "IPY_MODEL_78a65d554d0d4c3b8712c96391798a85" + ], + "layout": "IPY_MODEL_9f71e2c496e94633ab5cc7627713d0b9" + } + }, + "fc561d3e7239434f82e5f97834882373": { + "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_3c6ead09d8384533ad6d6dbc1e092bf7", + "placeholder": "​", + "style": "IPY_MODEL_7bb75c4743744fa4980ab172337295c6", + "value": "Loading weights: 100%" + } + }, + "39c70d8edbca4f71b171f8d278d3e7da": { + "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_06f346753e6341708bff1df229606ee7", + "max": 197, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b183d808dc054620bef47aa1abaa3b81", + "value": 197 + } + }, + "78a65d554d0d4c3b8712c96391798a85": { + "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_34a6088054c24d71b0a5301457f8feb6", + "placeholder": "​", + "style": "IPY_MODEL_0979317c03b248eda6f6a22b23201fdb", + "value": " 197/197 [00:00<00:00, 4340.38it/s, Materializing param=encoder.layer.11.output.dense.weight]" + } + }, + "9f71e2c496e94633ab5cc7627713d0b9": { + "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 + } + }, + "3c6ead09d8384533ad6d6dbc1e092bf7": { + "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 + } + }, + "7bb75c4743744fa4980ab172337295c6": { + "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": "" + } + }, + "06f346753e6341708bff1df229606ee7": { + "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 + } + }, + "b183d808dc054620bef47aa1abaa3b81": { + "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": "" + } + }, + "34a6088054c24d71b0a5301457f8feb6": { + "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 + } + }, + "0979317c03b248eda6f6a22b23201fdb": { + "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": "" + } + }, + "5848fb940af846cd933f3e422531bf8f": { + "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_838f04dc0d9542309882c1b3f4ade428", + "IPY_MODEL_304f366145bc45e2bb18a8cc8e580bd0", + "IPY_MODEL_d33504dcb688479d8bbbafc42a5c175b" + ], + "layout": "IPY_MODEL_15302e84f6b14b86be92850a8b64b2b2" + } + }, + "838f04dc0d9542309882c1b3f4ade428": { + "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_dae34e28a39d49da94c1f1423fc6f03e", + "placeholder": "​", + "style": "IPY_MODEL_95cb9b137b404a81a8a564d269aebfca", + "value": "tokenizer_config.json: 100%" + } + }, + "304f366145bc45e2bb18a8cc8e580bd0": { + "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_bbe461e321ed4f15bfb4a415ddaa6b5e", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_33e1e90c5b7d40539cb0d3eff1517215", + "value": 25 + } + }, + "d33504dcb688479d8bbbafc42a5c175b": { + "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_46a74a567a034bbfa01c1a38bf44a560", + "placeholder": "​", + "style": "IPY_MODEL_d3e26a070390464bbd887323979f6b9d", + "value": " 25.0/25.0 [00:00<00:00, 8.38kB/s]" + } + }, + "15302e84f6b14b86be92850a8b64b2b2": { + "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 + } + }, + "dae34e28a39d49da94c1f1423fc6f03e": { + "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 + } + }, + "95cb9b137b404a81a8a564d269aebfca": { + "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": "" + } + }, + "bbe461e321ed4f15bfb4a415ddaa6b5e": { + "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 + } + }, + "33e1e90c5b7d40539cb0d3eff1517215": { + "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": "" + } + }, + "46a74a567a034bbfa01c1a38bf44a560": { + "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 + } + }, + "d3e26a070390464bbd887323979f6b9d": { + "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": "" + } + }, + "59edb1de333f438b85b3a1c170ccbb99": { + "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_0254b5bac7ad4839893e165d49b99caa", + "IPY_MODEL_bd869600206c472f9c088e24cca59a06", + "IPY_MODEL_77c11e3cbdd344ea84f99dc532c74ac7" + ], + "layout": "IPY_MODEL_ca0c288dfb10404e8d3b006a1cd37547" + } + }, + "0254b5bac7ad4839893e165d49b99caa": { + "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_7756944b4f9b472f8dc77e65a0cb28b8", + "placeholder": "​", + "style": "IPY_MODEL_2a6eb8cfa49e4be6b63010060f0545d4", + "value": "vocab.json: " + } + }, + "bd869600206c472f9c088e24cca59a06": { + "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_4cb348841ab0465ab6020d0060446a8a", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_279f3fbfb3174fd7b6072776b3a18c17", + "value": 1 + } + }, + "77c11e3cbdd344ea84f99dc532c74ac7": { + "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_ce3f9808f34f484dabd51d98aa83c45c", + "placeholder": "​", + "style": "IPY_MODEL_6a13a1052fe7464493deab5f18087512", + "value": " 899k/? [00:00<00:00, 35.8MB/s]" + } + }, + "ca0c288dfb10404e8d3b006a1cd37547": { + "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 + } + }, + "7756944b4f9b472f8dc77e65a0cb28b8": { + "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 + } + }, + "2a6eb8cfa49e4be6b63010060f0545d4": { + "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": "" + } + }, + "4cb348841ab0465ab6020d0060446a8a": { + "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": "20px" + } + }, + "279f3fbfb3174fd7b6072776b3a18c17": { + "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": "" + } + }, + "ce3f9808f34f484dabd51d98aa83c45c": { + "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 + } + }, + "6a13a1052fe7464493deab5f18087512": { + "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": "" + } + }, + "97ae1151703b49c2b4552925c6bf146d": { + "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_dee89038279f4bf597b6fee92851846f", + "IPY_MODEL_8a859087e043457fb89e91c1cb7a7518", + "IPY_MODEL_36c55cdd3c594853b839a4fd804b6719" + ], + "layout": "IPY_MODEL_aa4b81df2e7c48e9a1fc58768ccab315" + } + }, + "dee89038279f4bf597b6fee92851846f": { + "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_531ef4668b0244379e244746a5663ac4", + "placeholder": "​", + "style": "IPY_MODEL_84101638e6d944dabf39d0fd107ebe8a", + "value": "merges.txt: " + } + }, + "8a859087e043457fb89e91c1cb7a7518": { + "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_7f7ac2e9e12648a1b13c92c11e159bd2", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6bfcbcb3ee5d4feb90366a3f392ed7de", + "value": 1 + } + }, + "36c55cdd3c594853b839a4fd804b6719": { + "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_1b7cee9b57c04a63b5dde8582963846f", + "placeholder": "​", + "style": "IPY_MODEL_b30bcdd589434fba8082f5c145ffba82", + "value": " 456k/? [00:00<00:00, 43.1MB/s]" + } + }, + "aa4b81df2e7c48e9a1fc58768ccab315": { + "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 + } + }, + "531ef4668b0244379e244746a5663ac4": { + "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 + } + }, + "84101638e6d944dabf39d0fd107ebe8a": { + "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": "" + } + }, + "7f7ac2e9e12648a1b13c92c11e159bd2": { + "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": "20px" + } + }, + "6bfcbcb3ee5d4feb90366a3f392ed7de": { + "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": "" + } + }, + "1b7cee9b57c04a63b5dde8582963846f": { + "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 + } + }, + "b30bcdd589434fba8082f5c145ffba82": { + "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": "" + } + }, + "5025ac720aaa4b0a99300be21c58a094": { + "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_47adec0f7ea64f9284262962d0ad5d67", + "IPY_MODEL_0496e5b26cfa4efead30f5f848bde313", + "IPY_MODEL_5f48f891cd7548d49115b6f87abe9ce2" + ], + "layout": "IPY_MODEL_ab79ef07a97c467494f92ad657acd890" + } + }, + "47adec0f7ea64f9284262962d0ad5d67": { + "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_7f2c27e564f6470285ffc1ad74a858ab", + "placeholder": "​", + "style": "IPY_MODEL_342f16b9972749178ffc62d3c5d6f0e5", + "value": "tokenizer.json: " + } + }, + "0496e5b26cfa4efead30f5f848bde313": { + "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_eae405f7ae774cfa952e8550f136ee50", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_1433f037c55246a2b5c62bfea9a7a287", + "value": 1 + } + }, + "5f48f891cd7548d49115b6f87abe9ce2": { + "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_6559e7091b1446239f8b320e1f5b3103", + "placeholder": "​", + "style": "IPY_MODEL_f4949533579247548a9ef75f12b4126a", + "value": " 1.36M/? [00:00<00:00, 149MB/s]" + } + }, + "ab79ef07a97c467494f92ad657acd890": { + "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 + } + }, + "7f2c27e564f6470285ffc1ad74a858ab": { + "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 + } + }, + "342f16b9972749178ffc62d3c5d6f0e5": { + "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": "" + } + }, + "eae405f7ae774cfa952e8550f136ee50": { + "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": "20px" + } + }, + "1433f037c55246a2b5c62bfea9a7a287": { + "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": "" + } + }, + "6559e7091b1446239f8b320e1f5b3103": { + "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 + } + }, + "f4949533579247548a9ef75f12b4126a": { + "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": "" + } + }, + "15cc16bacc074bc59293f5d824dcc8b4": { + "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_343fea9e802d40eb872f1234e4feeb9d", + "IPY_MODEL_bd17bb197fbd4081b1696427b9f4d37a", + "IPY_MODEL_82083d75226140ef9ffa0e568e355a7a" + ], + "layout": "IPY_MODEL_603a9a7cdd3c46518f110d0098513e92" + } + }, + "343fea9e802d40eb872f1234e4feeb9d": { + "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_bdab02de69de42239423c469597386df", + "placeholder": "​", + "style": "IPY_MODEL_e0e7903477c944ecb512ffe5f3ff54f1", + "value": "config.json: 100%" + } + }, + "bd17bb197fbd4081b1696427b9f4d37a": { + "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_7d453ba92fcb4873aa7f24fa56f0f4ff", + "max": 684, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_506ab6234b8a481ebf30a2457c47aba4", + "value": 684 + } + }, + "82083d75226140ef9ffa0e568e355a7a": { + "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_09b3f082f0234f44b448be363a8b5c4e", + "placeholder": "​", + "style": "IPY_MODEL_b6f08470c1cf45dd97d657092267efb9", + "value": " 684/684 [00:00<00:00, 205kB/s]" + } + }, + "603a9a7cdd3c46518f110d0098513e92": { + "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 + } + }, + "bdab02de69de42239423c469597386df": { + "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 + } + }, + "e0e7903477c944ecb512ffe5f3ff54f1": { + "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": "" + } + }, + "7d453ba92fcb4873aa7f24fa56f0f4ff": { + "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 + } + }, + "506ab6234b8a481ebf30a2457c47aba4": { + "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": "" + } + }, + "09b3f082f0234f44b448be363a8b5c4e": { + "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 + } + }, + "b6f08470c1cf45dd97d657092267efb9": { + "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": "" + } + }, + "e5bc5d3e749c4423b4fb9c3fc2328281": { + "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_9bcfee9c236f480487c6530f1a475b56", + "IPY_MODEL_2100916100ed468eacc8b746a6c9ba0b", + "IPY_MODEL_28717a14ebaa4708be6467202dc1cc72" + ], + "layout": "IPY_MODEL_149a12a56acf4ed5835bfb4e79ac9060" + } + }, + "9bcfee9c236f480487c6530f1a475b56": { + "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_e7c2e1cc77bf4a9f901453a987f50f0c", + "placeholder": "​", + "style": "IPY_MODEL_f8e5417887bb4815be3cf6c764fb18ef", + "value": "model.safetensors: 100%" + } + }, + "2100916100ed468eacc8b746a6c9ba0b": { + "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_6d87a649081d4a6db73dce63cc68fd05", + "max": 47372894, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e3fa86ff59f7459f9e8adf2414bcd0ba", + "value": 47372894 + } + }, + "28717a14ebaa4708be6467202dc1cc72": { + "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_faf9da9399ff43348ead176f92ceff68", + "placeholder": "​", + "style": "IPY_MODEL_971fbbb1e8c7475c99c0813eef1dd955", + "value": " 47.4M/47.4M [00:00<00:00, 236MB/s]" + } + }, + "149a12a56acf4ed5835bfb4e79ac9060": { + "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 + } + }, + "e7c2e1cc77bf4a9f901453a987f50f0c": { + "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 + } + }, + "f8e5417887bb4815be3cf6c764fb18ef": { + "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": "" + } + }, + "6d87a649081d4a6db73dce63cc68fd05": { + "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 + } + }, + "e3fa86ff59f7459f9e8adf2414bcd0ba": { + "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": "" + } + }, + "faf9da9399ff43348ead176f92ceff68": { + "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 + } + }, + "971fbbb1e8c7475c99c0813eef1dd955": { + "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": "" + } + }, + "98229bc34d8848e386f90b22ea5d0bde": { + "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_5b519c2321234664bec4fd9113022da3", + "IPY_MODEL_20bc786d1ce84edd843ee174d7c11edc", + "IPY_MODEL_24b48bc1221c45acab931780f51ef89f" + ], + "layout": "IPY_MODEL_0336c23eb7264b85b908f94a71564d14" + } + }, + "5b519c2321234664bec4fd9113022da3": { + "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_2126b3fd17364cf8b7c31b90b236300f", + "placeholder": "​", + "style": "IPY_MODEL_bd581a4138fe4eb4b1b98b5d13f9c6e4", + "value": "Loading weights: 100%" + } + }, + "20bc786d1ce84edd843ee174d7c11edc": { + "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_346c8a532cff426d8a815fdd87576481", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b051524ebc5e4913929a08198316452a", + "value": 25 + } + }, + "24b48bc1221c45acab931780f51ef89f": { + "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_10cc0a373a184a1a8557a663db2220e7", + "placeholder": "​", + "style": "IPY_MODEL_5a7102022fec4bc49e6d31bfc1474c24", + "value": " 25/25 [00:00<00:00, 2864.57it/s, Materializing param=pooler.weight]" + } + }, + "0336c23eb7264b85b908f94a71564d14": { + "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 + } + }, + "2126b3fd17364cf8b7c31b90b236300f": { + "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 + } + }, + "bd581a4138fe4eb4b1b98b5d13f9c6e4": { + "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": "" + } + }, + "346c8a532cff426d8a815fdd87576481": { + "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 + } + }, + "b051524ebc5e4913929a08198316452a": { + "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": "" + } + }, + "10cc0a373a184a1a8557a663db2220e7": { + "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 + } + }, + "5a7102022fec4bc49e6d31bfc1474c24": { + "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": "" + } + }, + "e128c3ccf1324bd5b6bc05b71aad92d2": { + "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_249baa0637614779ab3ea15097c1eea2", + "IPY_MODEL_d95ba50071bb434ca0e1fc5608e83d9b", + "IPY_MODEL_433a0f20e7194101bd56bb7a40ac4665" + ], + "layout": "IPY_MODEL_31f7c1ea6a5e4b41b7a37749958e172e" + } + }, + "249baa0637614779ab3ea15097c1eea2": { + "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_84a16620f70b4a0f91a90de18f460fd6", + "placeholder": "​", + "style": "IPY_MODEL_b8759548331345079689b26f35a79a77", + "value": "tokenizer_config.json: 100%" + } + }, + "d95ba50071bb434ca0e1fc5608e83d9b": { + "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_7910e302ff6f4cd8946644406e2c99ad", + "max": 25, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c0c3de1115a64039aeaa67ccc64324bb", + "value": 25 + } + }, + "433a0f20e7194101bd56bb7a40ac4665": { + "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_d2fb9021fdac446695c842d2c792bfa2", + "placeholder": "​", + "style": "IPY_MODEL_4a760f234dad418189cbf87761a8501d", + "value": " 25.0/25.0 [00:00<00:00, 9.29kB/s]" + } + }, + "31f7c1ea6a5e4b41b7a37749958e172e": { + "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 + } + }, + "84a16620f70b4a0f91a90de18f460fd6": { + "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 + } + }, + "b8759548331345079689b26f35a79a77": { + "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": "" + } + }, + "7910e302ff6f4cd8946644406e2c99ad": { + "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 + } + }, + "c0c3de1115a64039aeaa67ccc64324bb": { + "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": "" + } + }, + "d2fb9021fdac446695c842d2c792bfa2": { + "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 + } + }, + "4a760f234dad418189cbf87761a8501d": { + "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": "" + } + }, + "346fcb32c2d5444a9ad4a08ee14914f6": { + "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_0cfbbc11989c4dd7905f7091218759d7", + "IPY_MODEL_bfe3dde7bcf84031813d3f93c02a1779", + "IPY_MODEL_b2f4575b60d448f7b63144134ddf3d90" + ], + "layout": "IPY_MODEL_56380672ce2f4754aee7b7645644c0a5" + } + }, + "0cfbbc11989c4dd7905f7091218759d7": { + "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_89c87caf2e5e4e6db53132ecb99afe48", + "placeholder": "​", + "style": "IPY_MODEL_aa8f6c817dd941aa85621988025478cb", + "value": "spiece.model: 100%" + } + }, + "bfe3dde7bcf84031813d3f93c02a1779": { + "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_96d0554b09f0400eb9368c1a00026a10", + "max": 760289, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_1ff88b228a8e4c88beebb8c527db136d", + "value": 760289 + } + }, + "b2f4575b60d448f7b63144134ddf3d90": { + "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_33fab69359e04180bf754bf3e2ffaf1c", + "placeholder": "​", + "style": "IPY_MODEL_80b324a993bb458ea798bc61086ab048", + "value": " 760k/760k [00:00<00:00, 17.3MB/s]" + } + }, + "56380672ce2f4754aee7b7645644c0a5": { + "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 + } + }, + "89c87caf2e5e4e6db53132ecb99afe48": { + "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 + } + }, + "aa8f6c817dd941aa85621988025478cb": { + "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": "" + } + }, + "96d0554b09f0400eb9368c1a00026a10": { + "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 + } + }, + "1ff88b228a8e4c88beebb8c527db136d": { + "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": "" + } + }, + "33fab69359e04180bf754bf3e2ffaf1c": { + "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 + } + }, + "80b324a993bb458ea798bc61086ab048": { + "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": "" + } + }, + "b17e681239e44fa2a0ef188d3f4ed0b4": { + "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_9f072d8f80334c0bbbfd98fb7667ef8e", + "IPY_MODEL_db7afa36ff33456c841eb9e262bf64e4", + "IPY_MODEL_cbf45afff6d242fabce8f8c85ef356d4" + ], + "layout": "IPY_MODEL_af8f3e3993c247beb049f7208212cf71" + } + }, + "9f072d8f80334c0bbbfd98fb7667ef8e": { + "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_f0453d4dfdc64ff7bdb0b7b469347ece", + "placeholder": "​", + "style": "IPY_MODEL_e529b7ea2fc0414991532d82929eedb5", + "value": "tokenizer.json: " + } + }, + "db7afa36ff33456c841eb9e262bf64e4": { + "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_10ab5340121f40f0b7664651ea356c57", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ba77ceaa78c549238450c116edbde5a6", + "value": 1 + } + }, + "cbf45afff6d242fabce8f8c85ef356d4": { + "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_b9e1c8b7407143b89930decf94b4a0cf", + "placeholder": "​", + "style": "IPY_MODEL_4ef24dd03bf4481993c955959b436c7a", + "value": " 1.31M/? [00:00<00:00, 18.5MB/s]" + } + }, + "af8f3e3993c247beb049f7208212cf71": { + "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 + } + }, + "f0453d4dfdc64ff7bdb0b7b469347ece": { + "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 + } + }, + "e529b7ea2fc0414991532d82929eedb5": { + "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": "" + } + }, + "10ab5340121f40f0b7664651ea356c57": { + "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": "20px" + } + }, + "ba77ceaa78c549238450c116edbde5a6": { + "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": "" + } + }, + "b9e1c8b7407143b89930decf94b4a0cf": { + "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 + } + }, + "4ef24dd03bf4481993c955959b436c7a": { + "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": "" + } + }, + "3c5a3337721941078c409b79f532ee21": { + "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_8611029842bb498496be51f074148620", + "IPY_MODEL_18307a79e6934c979ff89cd62c6735f7", + "IPY_MODEL_d236e7740d6444598ca8fbc17a921aa5" + ], + "layout": "IPY_MODEL_746788a3733845539f86eccd5435af7a" + } + }, + "8611029842bb498496be51f074148620": { + "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_a93ded73f7324dcd8c35c5c941b8d376", + "placeholder": "​", + "style": "IPY_MODEL_c66a50a7121947a2aad757f17d02e50d", + "value": "config.json: 100%" + } + }, + "18307a79e6934c979ff89cd62c6735f7": { + "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_fd418721d3434d26b318a297d4f8a231", + "max": 483, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_64ff8bba2d6a42268d685e98dad5d3ba", + "value": 483 + } + }, + "d236e7740d6444598ca8fbc17a921aa5": { + "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_3b9094e941db4acabb47004acff6d5be", + "placeholder": "​", + "style": "IPY_MODEL_13879c1049c6437590de3c745eb66c10", + "value": " 483/483 [00:00<00:00, 112kB/s]" + } + }, + "746788a3733845539f86eccd5435af7a": { + "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 + } + }, + "a93ded73f7324dcd8c35c5c941b8d376": { + "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 + } + }, + "c66a50a7121947a2aad757f17d02e50d": { + "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": "" + } + }, + "fd418721d3434d26b318a297d4f8a231": { + "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 + } + }, + "64ff8bba2d6a42268d685e98dad5d3ba": { + "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": "" + } + }, + "3b9094e941db4acabb47004acff6d5be": { + "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 + } + }, + "13879c1049c6437590de3c745eb66c10": { + "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": "" + } + }, + "7e30d8c1c6b9477b85561ff45552f183": { + "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_da34458aa1cd451aa82bd0f1a201054e", + "IPY_MODEL_1e91cd8fd2a942a78f16e64e43647a68", + "IPY_MODEL_86e251461d0641cb816a6536a94d76b9" + ], + "layout": "IPY_MODEL_fe80e8d140cc41548750d8d0c033a745" + } + }, + "da34458aa1cd451aa82bd0f1a201054e": { + "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_31ba3c5b710f44a19a532311b6617e47", + "placeholder": "​", + "style": "IPY_MODEL_dfccd141316d446db864f9cee87d5134", + "value": "model.safetensors: 100%" + } + }, + "1e91cd8fd2a942a78f16e64e43647a68": { + "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_4996d3e41ade4d1aa77b8584cbd8cd90", + "max": 267954768, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_139887d421954ee08aca9e67f0ae1789", + "value": 267954768 + } + }, + "86e251461d0641cb816a6536a94d76b9": { + "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_767963f09cf1483fb94a3384e6c68393", + "placeholder": "​", + "style": "IPY_MODEL_55a1b96224d74308b4c5d3cfc54e6666", + "value": " 268M/268M [00:01<00:00, 292MB/s]" + } + }, + "fe80e8d140cc41548750d8d0c033a745": { + "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 + } + }, + "31ba3c5b710f44a19a532311b6617e47": { + "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 + } + }, + "dfccd141316d446db864f9cee87d5134": { + "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": "" + } + }, + "4996d3e41ade4d1aa77b8584cbd8cd90": { + "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 + } + }, + "139887d421954ee08aca9e67f0ae1789": { + "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": "" + } + }, + "767963f09cf1483fb94a3384e6c68393": { + "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 + } + }, + "55a1b96224d74308b4c5d3cfc54e6666": { + "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": "" + } + }, + "d7fd69d85bd94c1b9101de88917a56d0": { + "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_e9fe56efd05243c78bf3def4059cc5d5", + "IPY_MODEL_d402754d2c214f129008730d035dfb3e", + "IPY_MODEL_edb0d48ba263454297e61976eb78669b" + ], + "layout": "IPY_MODEL_3f0843e7b6f64d4bb6f1787f27f2b5b3" + } + }, + "e9fe56efd05243c78bf3def4059cc5d5": { + "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_9ae9ac7e39294492b02429732a76200a", + "placeholder": "​", + "style": "IPY_MODEL_c72181fa23a9422298c8955dd7be1f94", + "value": "Loading weights: 100%" + } + }, + "d402754d2c214f129008730d035dfb3e": { + "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_16abbacb22d44575a9141cbcb14b6fad", + "max": 100, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_bc9c008333e34c8fa5cda6067e91e306", + "value": 100 + } + }, + "edb0d48ba263454297e61976eb78669b": { + "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_c91a2fb1f14244aeb9da258ccd0636ce", + "placeholder": "​", + "style": "IPY_MODEL_84372dd446174ad8803d1cd7698c47e3", + "value": " 100/100 [00:00<00:00, 3628.42it/s, Materializing param=transformer.layer.5.sa_layer_norm.weight]" + } + }, + "3f0843e7b6f64d4bb6f1787f27f2b5b3": { + "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 + } + }, + "9ae9ac7e39294492b02429732a76200a": { + "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 + } + }, + "c72181fa23a9422298c8955dd7be1f94": { + "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": "" + } + }, + "16abbacb22d44575a9141cbcb14b6fad": { + "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 + } + }, + "bc9c008333e34c8fa5cda6067e91e306": { + "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": "" + } + }, + "c91a2fb1f14244aeb9da258ccd0636ce": { + "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 + } + }, + "84372dd446174ad8803d1cd7698c47e3": { + "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": "" + } + }, + "d5ac8b79cdb74d11ae1cea4a253a7813": { + "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_48171f3cdbc945689fa3c99a88efc262", + "IPY_MODEL_a74b89ef8f4c4fe4b92160fa46a3eeb3", + "IPY_MODEL_c96fdd2be2a84390b96eb2b20fbf2231" + ], + "layout": "IPY_MODEL_aa15c5e6a8ac4d05862ce73125d7c40f" + } + }, + "48171f3cdbc945689fa3c99a88efc262": { + "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_b8ab4b12a6fb4887bb6c0333cbd91ea8", + "placeholder": "​", + "style": "IPY_MODEL_0b6b4d4182eb46dcb9bd0a5eca5f60ee", + "value": "tokenizer_config.json: 100%" + } + }, + "a74b89ef8f4c4fe4b92160fa46a3eeb3": { + "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_c8f02de241004227ba590a1432ea6e84", + "max": 48, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_70e78c654d484aa9ab406af8ba91ab0d", + "value": 48 + } + }, + "c96fdd2be2a84390b96eb2b20fbf2231": { + "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_b918dc885abb450dba336c502b97b861", + "placeholder": "​", + "style": "IPY_MODEL_932896d6d10444d4a159306cb31d8d6a", + "value": " 48.0/48.0 [00:00<00:00, 15.0kB/s]" + } + }, + "aa15c5e6a8ac4d05862ce73125d7c40f": { + "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 + } + }, + "b8ab4b12a6fb4887bb6c0333cbd91ea8": { + "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 + } + }, + "0b6b4d4182eb46dcb9bd0a5eca5f60ee": { + "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": "" + } + }, + "c8f02de241004227ba590a1432ea6e84": { + "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 + } + }, + "70e78c654d484aa9ab406af8ba91ab0d": { + "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": "" + } + }, + "b918dc885abb450dba336c502b97b861": { + "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 + } + }, + "932896d6d10444d4a159306cb31d8d6a": { + "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": "" + } + }, + "93a1ce256ee440ed9084902937a5621d": { + "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_d0f3d668b5dc40dda90b6c845c2ef7b4", + "IPY_MODEL_1650dfc316274bce8ecf6c071c687f1e", + "IPY_MODEL_d928f6cb5fc943129ae571a421b7a2d4" + ], + "layout": "IPY_MODEL_7881a6e94d8c4dc8bfe375bedc5eafe7" + } + }, + "d0f3d668b5dc40dda90b6c845c2ef7b4": { + "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_eb3abb7ba6d64b94b65f27177c46b411", + "placeholder": "​", + "style": "IPY_MODEL_f3acb74c2c594020aedf6330ee1f1847", + "value": "vocab.txt: " + } + }, + "1650dfc316274bce8ecf6c071c687f1e": { + "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_b83c083a5acf458a9f4ed93f5037b84b", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_615e50faa3e34a3380f2012a65fa3825", + "value": 1 + } + }, + "d928f6cb5fc943129ae571a421b7a2d4": { + "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_4a652972f8084b9caaf798d2f99ca6b7", + "placeholder": "​", + "style": "IPY_MODEL_a611df0fed2242d586dea0adb1197b44", + "value": " 232k/? [00:00<00:00, 12.3MB/s]" + } + }, + "7881a6e94d8c4dc8bfe375bedc5eafe7": { + "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 + } + }, + "eb3abb7ba6d64b94b65f27177c46b411": { + "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 + } + }, + "f3acb74c2c594020aedf6330ee1f1847": { + "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": "" + } + }, + "b83c083a5acf458a9f4ed93f5037b84b": { + "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": "20px" + } + }, + "615e50faa3e34a3380f2012a65fa3825": { + "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": "" + } + }, + "4a652972f8084b9caaf798d2f99ca6b7": { + "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 + } + }, + "a611df0fed2242d586dea0adb1197b44": { + "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": "" + } + }, + "e8336e9cea94474e9874959ca69d81f7": { + "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_802b8b73a4074d5691ed180a879cbdcc", + "IPY_MODEL_cbee27259204437d873d80ab8b70932f", + "IPY_MODEL_a19ec4ba40fd46548f12c4a5b11fbfdc" + ], + "layout": "IPY_MODEL_d2abb6d0e55a4b9ab027964c19291751" + } + }, + "802b8b73a4074d5691ed180a879cbdcc": { + "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_a738f491166d4e90820c6f88f2038cf6", + "placeholder": "​", + "style": "IPY_MODEL_d5b57aea91f84665bbc41dbb4dda3230", + "value": "tokenizer.json: " + } + }, + "cbee27259204437d873d80ab8b70932f": { + "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_4d62d0426c3649fab4c374d0fe33dbea", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5351971c88e3442fba201147a6303fa2", + "value": 1 + } + }, + "a19ec4ba40fd46548f12c4a5b11fbfdc": { + "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_6ecdc158d0b442e0bb46ddc3909c5039", + "placeholder": "​", + "style": "IPY_MODEL_c13b289614424c148473c823e42b854a", + "value": " 466k/? [00:00<00:00, 59.4MB/s]" + } + }, + "d2abb6d0e55a4b9ab027964c19291751": { + "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 + } + }, + "a738f491166d4e90820c6f88f2038cf6": { + "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 + } + }, + "d5b57aea91f84665bbc41dbb4dda3230": { + "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": "" + } + }, + "4d62d0426c3649fab4c374d0fe33dbea": { + "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": "20px" + } + }, + "5351971c88e3442fba201147a6303fa2": { + "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": "" + } + }, + "6ecdc158d0b442e0bb46ddc3909c5039": { + "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 + } + }, + "c13b289614424c148473c823e42b854a": { + "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": "" + } + }, + "a0977fe1e5cf409fb3513fa65a83da47": { + "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_dd049390fe7e49f0849e5cf1de250b21", + "IPY_MODEL_65a93ea8237c458dab18cd8fcc3783c1", + "IPY_MODEL_d4e13efd20be41a982c0eab2bc1e50d7" + ], + "layout": "IPY_MODEL_cd4f229665d047f6a49ad9b2f7154502" + } + }, + "dd049390fe7e49f0849e5cf1de250b21": { + "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_d946f3a09f9b4f659a85368d25145aa6", + "placeholder": "​", + "style": "IPY_MODEL_546b70c09f344f1f909c1e12e2c4e51b", + "value": "Loading weights: 100%" + } + }, + "65a93ea8237c458dab18cd8fcc3783c1": { + "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_0487417b6f1b4b7e8ebba263c895bf0b", + "max": 82, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_f3749087986a4a548f59131c1c74df6f", + "value": 82 + } + }, + "d4e13efd20be41a982c0eab2bc1e50d7": { + "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_03e2b8789387474d92d449a1946dfceb", + "placeholder": "​", + "style": "IPY_MODEL_a96fd2023e2d45848a631e22fb05e028", + "value": " 82/82 [00:00<00:00, 3652.45it/s, Materializing param=token_emb.weight]" + } + }, + "cd4f229665d047f6a49ad9b2f7154502": { + "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 + } + }, + "d946f3a09f9b4f659a85368d25145aa6": { + "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 + } + }, + "546b70c09f344f1f909c1e12e2c4e51b": { + "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": "" + } + }, + "0487417b6f1b4b7e8ebba263c895bf0b": { + "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 + } + }, + "f3749087986a4a548f59131c1c74df6f": { + "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": "" + } + }, + "03e2b8789387474d92d449a1946dfceb": { + "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 + } + }, + "a96fd2023e2d45848a631e22fb05e028": { + "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": "" + } + }, + "844ca279a1214e9aaf01e0ce607f597b": { + "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_6f81df77f8754e4984edee91b95c06f6", + "IPY_MODEL_5c2eccc71ebd4a419dc32698b1760fd0", + "IPY_MODEL_45e4090dafd94747bd871805029e608e" + ], + "layout": "IPY_MODEL_cb0a41a2ef4f4cdfa6d1d1a7f924002c" + } + }, + "6f81df77f8754e4984edee91b95c06f6": { + "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_8cb6b1dfd31146419ff33c919f2901cc", + "placeholder": "​", + "style": "IPY_MODEL_9fb532dda749453c89c8ab044ad31529", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "5c2eccc71ebd4a419dc32698b1760fd0": { + "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_78d12e22baad475f947fcda09b5278b8", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_a04c0b91b85945bb834baf72f4183b6c", + "value": 1 + } + }, + "45e4090dafd94747bd871805029e608e": { + "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_d240e19d83f5499c8694e809fff45fac", + "placeholder": "​", + "style": "IPY_MODEL_2d630d5683194818bf6cdfbc927c780b", + "value": "  130MB /  130MB, 56.3MB/s  " + } + }, + "cb0a41a2ef4f4cdfa6d1d1a7f924002c": { + "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 + } + }, + "8cb6b1dfd31146419ff33c919f2901cc": { + "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 + } + }, + "9fb532dda749453c89c8ab044ad31529": { + "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": "" + } + }, + "78d12e22baad475f947fcda09b5278b8": { + "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": "20px" + } + }, + "a04c0b91b85945bb834baf72f4183b6c": { + "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": "" + } + }, + "d240e19d83f5499c8694e809fff45fac": { + "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 + } + }, + "2d630d5683194818bf6cdfbc927c780b": { + "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": "" + } + }, + "daabbc1ccb994d4c91e49278ad8fc7e9": { + "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_abdd527ee4294e6f9c6fa656f56105cc", + "IPY_MODEL_ddba119e49ae4e4ba80a19d42695652c", + "IPY_MODEL_3628a737a8154beeb0acb3793a396dfb" + ], + "layout": "IPY_MODEL_0ad5fae36ddc4977afcf00ac6be6d468" + } + }, + "abdd527ee4294e6f9c6fa656f56105cc": { + "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_63a7abab53aa4f489f0bda1babc0d408", + "placeholder": "​", + "style": "IPY_MODEL_dc9d9cbf8c27457f863c5fd43c97492a", + "value": "New Data Upload               : " + } + }, + "ddba119e49ae4e4ba80a19d42695652c": { + "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_f831e6630b7147528886e12888898e67", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_772345ef17ea42728007a6edc31f418e", + "value": 0 + } + }, + "3628a737a8154beeb0acb3793a396dfb": { + "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_630e9a7c293d41bfa126a101ce0cf63f", + "placeholder": "​", + "style": "IPY_MODEL_eb079b0bb2224406baf26f0d63e9fcb0", + "value": "  0.00B /  0.00B,  0.00B/s  " + } + }, + "0ad5fae36ddc4977afcf00ac6be6d468": { + "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 + } + }, + "63a7abab53aa4f489f0bda1babc0d408": { + "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 + } + }, + "dc9d9cbf8c27457f863c5fd43c97492a": { + "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": "" + } + }, + "f831e6630b7147528886e12888898e67": { + "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": "20px" + } + }, + "772345ef17ea42728007a6edc31f418e": { + "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": "" + } + }, + "630e9a7c293d41bfa126a101ce0cf63f": { + "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 + } + }, + "eb079b0bb2224406baf26f0d63e9fcb0": { + "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": "" + } + }, + "f9b8c0e168e04a1b91aac96442e45c4b": { + "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_d7d5ac10cb5045c387abf2831e52d066", + "IPY_MODEL_746c4209461a44308fae532dc1fe3718", + "IPY_MODEL_9f667e0f66df420b906359b35aba546e" + ], + "layout": "IPY_MODEL_909ee967105846e8845df4d541951ae2" + } + }, + "d7d5ac10cb5045c387abf2831e52d066": { + "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_0fa661e22316455aae493914ec79951d", + "placeholder": "​", + "style": "IPY_MODEL_79ed5fc5b4264a1a97f472045c7da4b1", + "value": "  /tmp/model.safetensors      : 100%" + } + }, + "746c4209461a44308fae532dc1fe3718": { + "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_aa3af48f96c94cc0909e2a4cf2c81dca", + "max": 129711020, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ec9757b004e24e599d8cea252eb514a5", + "value": 129711020 + } + }, + "9f667e0f66df420b906359b35aba546e": { + "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_ba2a9dad912746b1bc068c65253f9516", + "placeholder": "​", + "style": "IPY_MODEL_dd1fd477b0584b9895207e6764f2aad7", + "value": "  130MB /  130MB            " + } + }, + "909ee967105846e8845df4d541951ae2": { + "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 + } + }, + "0fa661e22316455aae493914ec79951d": { + "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 + } + }, + "79ed5fc5b4264a1a97f472045c7da4b1": { + "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": "" + } + }, + "aa3af48f96c94cc0909e2a4cf2c81dca": { + "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 + } + }, + "ec9757b004e24e599d8cea252eb514a5": { + "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": "" + } + }, + "ba2a9dad912746b1bc068c65253f9516": { + "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 + } + }, + "dd1fd477b0584b9895207e6764f2aad7": { + "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": "" + } + }, + "9a5bcab81f4f4e6d9d0cf0dc0ad7f072": { + "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_f2844a1d8cd44bb5a319637dfc22bdf0", + "IPY_MODEL_5448afee2fb246f89f6cf11bd5dccbc8", + "IPY_MODEL_73427526fc7d4af4aae82a12183f34fb" + ], + "layout": "IPY_MODEL_9a343c75782a415580509fdf2d56b581" + } + }, + "f2844a1d8cd44bb5a319637dfc22bdf0": { + "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_02998e1a96de4f92a5b6a66ae3d73c14", + "placeholder": "​", + "style": "IPY_MODEL_4e587115d78f4964b0f380ab3609f072", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "5448afee2fb246f89f6cf11bd5dccbc8": { + "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_1b75e405ffc04d3499b5d2767e599ae8", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_1d03959538e3448da85423c192df8e02", + "value": 1 + } + }, + "73427526fc7d4af4aae82a12183f34fb": { + "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_9967f0d360954a88b38a5ebe4e1bc450", + "placeholder": "​", + "style": "IPY_MODEL_3071e444c3694300858adcc023949187", + "value": " 25.9MB / 25.9MB, 1.28MB/s  " + } + }, + "9a343c75782a415580509fdf2d56b581": { + "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 + } + }, + "02998e1a96de4f92a5b6a66ae3d73c14": { + "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 + } + }, + "4e587115d78f4964b0f380ab3609f072": { + "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": "" + } + }, + "1b75e405ffc04d3499b5d2767e599ae8": { + "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": "20px" + } + }, + "1d03959538e3448da85423c192df8e02": { + "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": "" + } + }, + "9967f0d360954a88b38a5ebe4e1bc450": { + "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 + } + }, + "3071e444c3694300858adcc023949187": { + "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": "" + } + }, + "f4f687f2230b42a487b76a0c053d7d3c": { + "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_7b0a7b4dacbd431ea1d9ede26840efd5", + "IPY_MODEL_f2ab7b1fd39c40258f361be283a1be71", + "IPY_MODEL_65c0717a4c264140a8f519f9a739dabb" + ], + "layout": "IPY_MODEL_5ddd5a7845f74cb2a215e8d00894cd43" + } + }, + "7b0a7b4dacbd431ea1d9ede26840efd5": { + "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_92f57381b25d4c01b573db158a49a21c", + "placeholder": "​", + "style": "IPY_MODEL_28da595390514fff8f3128b6d7b8788f", + "value": "New Data Upload      ��        : 100%" + } + }, + "f2ab7b1fd39c40258f361be283a1be71": { + "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_03f4a1bb4ffe4c7a86db79d1a28f68a9", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_57022bfea9a645a5a0081f6c1030c19d", + "value": 1 + } + }, + "65c0717a4c264140a8f519f9a739dabb": { + "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_68d4e310cf56489e94bd7ec9df3697ce", + "placeholder": "​", + "style": "IPY_MODEL_2dc73c08adbc42c1959c7d5a8e5ae6e2", + "value": " 1.35MB / 1.35MB, 1.28MB/s  " + } + }, + "5ddd5a7845f74cb2a215e8d00894cd43": { + "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 + } + }, + "92f57381b25d4c01b573db158a49a21c": { + "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 + } + }, + "28da595390514fff8f3128b6d7b8788f": { + "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": "" + } + }, + "03f4a1bb4ffe4c7a86db79d1a28f68a9": { + "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": "20px" + } + }, + "57022bfea9a645a5a0081f6c1030c19d": { + "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": "" + } + }, + "68d4e310cf56489e94bd7ec9df3697ce": { + "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 + } + }, + "2dc73c08adbc42c1959c7d5a8e5ae6e2": { + "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": "" + } + }, + "be0e1a5f9c6c4209b8aa90a184d271dc": { + "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_540cdd218e2e40a0b8860ab4d23d75e7", + "IPY_MODEL_abd5346d99014ddca5d9a9b862e6c171", + "IPY_MODEL_de3186bec27e453d870688404cb7569f" + ], + "layout": "IPY_MODEL_993e8da4216b438da776edb360ffe123" + } + }, + "540cdd218e2e40a0b8860ab4d23d75e7": { + "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_330886fd8b7741a49e67c2589dc0d7cc", + "placeholder": "​", + "style": "IPY_MODEL_781271b67f134a288cea9ed362373788", + "value": "  alignment_bank_best.pt      : 100%" + } + }, + "abd5346d99014ddca5d9a9b862e6c171": { + "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_3a4d04ca69374a81b35dc371005945cb", + "max": 25878457, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_3e26f6f61a134f73ade5d754344f78ba", + "value": 25878457 + } + }, + "de3186bec27e453d870688404cb7569f": { + "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_58b0317a4b724a96a67e200820aea5b4", + "placeholder": "​", + "style": "IPY_MODEL_8821c2d84a4f4eb4afd3d886b60d1d29", + "value": " 25.9MB / 25.9MB            " + } + }, + "993e8da4216b438da776edb360ffe123": { + "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 + } + }, + "330886fd8b7741a49e67c2589dc0d7cc": { + "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 + } + }, + "781271b67f134a288cea9ed362373788": { + "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": "" + } + }, + "3a4d04ca69374a81b35dc371005945cb": { + "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 + } + }, + "3e26f6f61a134f73ade5d754344f78ba": { + "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": "" + } + }, + "58b0317a4b724a96a67e200820aea5b4": { + "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 + } + }, + "8821c2d84a4f4eb4afd3d886b60d1d29": { + "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": "" + } + }, + "aafdfd194a344d46805476fe892af3a1": { + "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_531ab6c2d71145e8adc057385f7817bc", + "IPY_MODEL_a9bff173bdac4e24ad18922d0ba619d5", + "IPY_MODEL_247c1842cd734846bfc682156a6d7156" + ], + "layout": "IPY_MODEL_dae98cdc7e644e678d56705284f33367" + } + }, + "531ab6c2d71145e8adc057385f7817bc": { + "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_df405bcdfc384f7d910ab7d5062e0999", + "placeholder": "​", + "style": "IPY_MODEL_a5ae260280294bf8be1de758ef7e1149", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "a9bff173bdac4e24ad18922d0ba619d5": { + "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_79ec96f0a7914deba06ac738d4c30600", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_0675c64897d14cf4afe5f655cefb27ca", + "value": 1 + } + }, + "247c1842cd734846bfc682156a6d7156": { + "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_2c37bec7a7104904bca4bdf312f29da0", + "placeholder": "​", + "style": "IPY_MODEL_8f45e2954ee54f0e9fad63f70a85667d", + "value": " 25.9MB / 25.9MB, 17.7MB/s  " + } + }, + "dae98cdc7e644e678d56705284f33367": { + "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 + } + }, + "df405bcdfc384f7d910ab7d5062e0999": { + "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 + } + }, + "a5ae260280294bf8be1de758ef7e1149": { + "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": "" + } + }, + "79ec96f0a7914deba06ac738d4c30600": { + "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": "20px" + } + }, + "0675c64897d14cf4afe5f655cefb27ca": { + "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": "" + } + }, + "2c37bec7a7104904bca4bdf312f29da0": { + "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 + } + }, + "8f45e2954ee54f0e9fad63f70a85667d": { + "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": "" + } + }, + "06aa2181f5b14f91b7e1c916564c0f1b": { + "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_9b619359df054bcc9cac920733bac1f2", + "IPY_MODEL_278a85927a56473f9eb31fd950522917", + "IPY_MODEL_44ac24f7a99045489434c8961453d280" + ], + "layout": "IPY_MODEL_8730284d606a478492c00fa762a745a3" + } + }, + "9b619359df054bcc9cac920733bac1f2": { + "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_3e5d8372c100477294f1155c7e55a08e", + "placeholder": "​", + "style": "IPY_MODEL_5750e73b7271438cbed9e6a413641e88", + "value": "New Data Upload               : 100%" + } + }, + "278a85927a56473f9eb31fd950522917": { + "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_bc77094115744d97b78821775ec18f56", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_fdcbc4447f7b42388dad192c54627b58", + "value": 1 + } + }, + "44ac24f7a99045489434c8961453d280": { + "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_d581bdec81734d68bd5dd7af74f345ff", + "placeholder": "​", + "style": "IPY_MODEL_819362f51b7745328e8a3b3be5ad1290", + "value": " 25.9MB / 25.9MB, 17.7MB/s  " + } + }, + "8730284d606a478492c00fa762a745a3": { + "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 + } + }, + "3e5d8372c100477294f1155c7e55a08e": { + "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 + } + }, + "5750e73b7271438cbed9e6a413641e88": { + "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": "" + } + }, + "bc77094115744d97b78821775ec18f56": { + "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": "20px" + } + }, + "fdcbc4447f7b42388dad192c54627b58": { + "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": "" + } + }, + "d581bdec81734d68bd5dd7af74f345ff": { + "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 + } + }, + "819362f51b7745328e8a3b3be5ad1290": { + "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": "" + } + }, + "aec73061149c42f8a8f9ecd1f5040657": { + "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_a3c2ddcb69dd40bf9289e090d8c235cd", + "IPY_MODEL_0a160f09ddfb40bf983fda72038dcff1", + "IPY_MODEL_034640b2d7bf450c837f3b2a996be2ff" + ], + "layout": "IPY_MODEL_f429d9eedc8f4495b11534533fa7d1a4" + } + }, + "a3c2ddcb69dd40bf9289e090d8c235cd": { + "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_ff95002595e84e8a805ec6f22fd34677", + "placeholder": "​", + "style": "IPY_MODEL_d43594d8e7884c4ab2b7ccac1f1c0227", + "value": "  alignment_bank_final.pt     : 100%" + } + }, + "0a160f09ddfb40bf983fda72038dcff1": { + "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_72ccbf31bee9462283e8ac9aad8d529f", + "max": 25878493, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d954c2064cba42f8b561a6f067fbf53b", + "value": 25878493 + } + }, + "034640b2d7bf450c837f3b2a996be2ff": { + "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_36c6a171fa4c4dfc87a121243d90cef4", + "placeholder": "​", + "style": "IPY_MODEL_2bff46e20c084c4798dd9012950d6bd5", + "value": " 25.9MB / 25.9MB            " + } + }, + "f429d9eedc8f4495b11534533fa7d1a4": { + "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 + } + }, + "ff95002595e84e8a805ec6f22fd34677": { + "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 + } + }, + "d43594d8e7884c4ab2b7ccac1f1c0227": { + "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": "" + } + }, + "72ccbf31bee9462283e8ac9aad8d529f": { + "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 + } + }, + "d954c2064cba42f8b561a6f067fbf53b": { + "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": "" + } + }, + "36c6a171fa4c4dfc87a121243d90cef4": { + "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 + } + }, + "2bff46e20c084c4798dd9012950d6bd5": { + "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": "" + } + }, + "3ea7d1536b1a487dbeb3200dca65f573": { + "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_a4b28207be614284b4da30bd1c243ab7", + "IPY_MODEL_cc42746c36b04efe9d0019516ab15b24", + "IPY_MODEL_89d355da1e4245d9b8c73f58f45e16ac" + ], + "layout": "IPY_MODEL_475c3ee99a814b41befb04df1a0dcb1b" + } + }, + "a4b28207be614284b4da30bd1c243ab7": { + "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_47deda8ea4464f9fb573ce607ec0b457", + "placeholder": "​", + "style": "IPY_MODEL_5b3fc4dd048a44cc802de7d36a554840", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "cc42746c36b04efe9d0019516ab15b24": { + "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_9ba7853e89064fc0baf96032bef06ea7", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_70536b254fec4142b3e04a9b8598836f", + "value": 1 + } + }, + "89d355da1e4245d9b8c73f58f45e16ac": { + "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_9db2c00ace564417b68231cf51486282", + "placeholder": "​", + "style": "IPY_MODEL_9e85c55583a9464992c6eca4bdf90b91", + "value": "  104MB /  104MB, 19.8MB/s  " + } + }, + "475c3ee99a814b41befb04df1a0dcb1b": { + "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 + } + }, + "47deda8ea4464f9fb573ce607ec0b457": { + "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 + } + }, + "5b3fc4dd048a44cc802de7d36a554840": { + "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": "" + } + }, + "9ba7853e89064fc0baf96032bef06ea7": { + "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": "20px" + } + }, + "70536b254fec4142b3e04a9b8598836f": { + "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": "" + } + }, + "9db2c00ace564417b68231cf51486282": { + "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 + } + }, + "9e85c55583a9464992c6eca4bdf90b91": { + "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": "" + } + }, + "b952d107e80d4ddeb9f61397e3968659": { + "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_06b2f006b1ab48359a36b91a21b48e87", + "IPY_MODEL_3983fbc28cf8447e872e9495714905e1", + "IPY_MODEL_a1c68a4910944d4cbaa582d6621497a3" + ], + "layout": "IPY_MODEL_040ea1a742df41d1b8ba7754bf93230d" + } + }, + "06b2f006b1ab48359a36b91a21b48e87": { + "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_c76fee1c9ae74473b2aa32ae8a030de6", + "placeholder": "​", + "style": "IPY_MODEL_afd6158d26e1491ca4cffcbb8cab0c8f", + "value": "New Data Upload               : " + } + }, + "3983fbc28cf8447e872e9495714905e1": { + "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_6d08deefc3fe43208ecdd551cdd670a1", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_30d525a65572435292c59fa8c60347d4", + "value": 0 + } + }, + "a1c68a4910944d4cbaa582d6621497a3": { + "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_71c8482e69594de7841308c4afb8ddb6", + "placeholder": "​", + "style": "IPY_MODEL_0a4e2574d38d4f3eb84297c4ae99cf93", + "value": "  0.00B /  0.00B,  0.00B/s  " + } + }, + "040ea1a742df41d1b8ba7754bf93230d": { + "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 + } + }, + "c76fee1c9ae74473b2aa32ae8a030de6": { + "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 + } + }, + "afd6158d26e1491ca4cffcbb8cab0c8f": { + "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": "" + } + }, + "6d08deefc3fe43208ecdd551cdd670a1": { + "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": "20px" + } + }, + "30d525a65572435292c59fa8c60347d4": { + "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": "" + } + }, + "71c8482e69594de7841308c4afb8ddb6": { + "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 + } + }, + "0a4e2574d38d4f3eb84297c4ae99cf93": { + "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": "" + } + }, + "5570f9abda9747d8b81e459efc6f420d": { + "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_ac7892cc541e4a858db94a82b5ffdf46", + "IPY_MODEL_3954a8f50e8348419301305de7637203", + "IPY_MODEL_36a1da91b520487cb00bb3ae9f68322a" + ], + "layout": "IPY_MODEL_db412f7b422f443ab9fc02f5c8a7267a" + } + }, + "ac7892cc541e4a858db94a82b5ffdf46": { + "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_5852e1611ff646b587dcd59d2c2cd7fa", + "placeholder": "​", + "style": "IPY_MODEL_63af04e265734e4ea0edbabc3637ac77", + "value": "  ...3d88d5e6e1a/best_model.pt: 100%" + } + }, + "3954a8f50e8348419301305de7637203": { + "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_722d6f4a94ab4270b4bd8438e1a1b2d3", + "max": 103862453, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_bc4f89591ebc47298d8d1732dbdc70ef", + "value": 103862453 + } + }, + "36a1da91b520487cb00bb3ae9f68322a": { + "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_f8f2092ad53649019835aec315d2ad6d", + "placeholder": "​", + "style": "IPY_MODEL_8e2755a19c3f418fa777030efd951486", + "value": "  104MB /  104MB            " + } + }, + "db412f7b422f443ab9fc02f5c8a7267a": { + "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 + } + }, + "5852e1611ff646b587dcd59d2c2cd7fa": { + "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 + } + }, + "63af04e265734e4ea0edbabc3637ac77": { + "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": "" + } + }, + "722d6f4a94ab4270b4bd8438e1a1b2d3": { + "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 + } + }, + "bc4f89591ebc47298d8d1732dbdc70ef": { + "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": "" + } + }, + "f8f2092ad53649019835aec315d2ad6d": { + "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 + } + }, + "8e2755a19c3f418fa777030efd951486": { + "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": "" + } + }, + "b84fd89752544a1bab4f039217c52d83": { + "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_f04c917e1ffa4918ad00ecbca177741b", + "IPY_MODEL_0bb086c92ef7429f9ab48c89e4f847fc", + "IPY_MODEL_57b0ec87b7ae41bdaf3cc565c6612de8" + ], + "layout": "IPY_MODEL_9428c5c782384f139b301d0e4146c475" + } + }, + "f04c917e1ffa4918ad00ecbca177741b": { + "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_249e7f3721c74243bf59cdb3b65f2350", + "placeholder": "​", + "style": "IPY_MODEL_e320cb11d8cc44638fc997e4bc3066b6", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "0bb086c92ef7429f9ab48c89e4f847fc": { + "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_1b0b0e4642b5444eb7188f934a1a45a9", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7ae63114504d421283e7bd5df8fe5c2d", + "value": 1 + } + }, + "57b0ec87b7ae41bdaf3cc565c6612de8": { + "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_b20ca3a889f249c5a4ec61faab39f3b9", + "placeholder": "​", + "style": "IPY_MODEL_00b6693fedc24dd4b6018b1d7ad66a22", + "value": " 1.54GB / 1.54GB, 15.3MB/s  " + } + }, + "9428c5c782384f139b301d0e4146c475": { + "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 + } + }, + "249e7f3721c74243bf59cdb3b65f2350": { + "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 + } + }, + "e320cb11d8cc44638fc997e4bc3066b6": { + "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": "" + } + }, + "1b0b0e4642b5444eb7188f934a1a45a9": { + "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": "20px" + } + }, + "7ae63114504d421283e7bd5df8fe5c2d": { + "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": "" + } + }, + "b20ca3a889f249c5a4ec61faab39f3b9": { + "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 + } + }, + "00b6693fedc24dd4b6018b1d7ad66a22": { + "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": "" + } + }, + "7f33873d36b045ddbf4cebab46902c8b": { + "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_142260d1a68e419981476cf344fede3d", + "IPY_MODEL_01a38c16c788493ca07e128f96afae89", + "IPY_MODEL_0f8126ee7e9a4740a4f084ddc716a1dd" + ], + "layout": "IPY_MODEL_3fb95746b95540178f0e3037429fc60c" + } + }, + "142260d1a68e419981476cf344fede3d": { + "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_28a5c117317c43388b93b8550b78d4bf", + "placeholder": "​", + "style": "IPY_MODEL_1210f005a3b94c9ea34b198213a809be", + "value": "New Data Upload               : 100%" + } + }, + "01a38c16c788493ca07e128f96afae89": { + "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_e056dfcd0ed64eacb4b4bafa5acf5ac8", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5ab6f5b71c1441509fe834f0bf42cc58", + "value": 1 + } + }, + "0f8126ee7e9a4740a4f084ddc716a1dd": { + "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_da55e960029542bf97e9b6190471f818", + "placeholder": "​", + "style": "IPY_MODEL_9e4dbbdf7d694ce491c2bad345e437be", + "value": " 1.54GB / 1.54GB, 15.3MB/s  " + } + }, + "3fb95746b95540178f0e3037429fc60c": { + "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 + } + }, + "28a5c117317c43388b93b8550b78d4bf": { + "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 + } + }, + "1210f005a3b94c9ea34b198213a809be": { + "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": "" + } + }, + "e056dfcd0ed64eacb4b4bafa5acf5ac8": { + "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": "20px" + } + }, + "5ab6f5b71c1441509fe834f0bf42cc58": { + "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": "" + } + }, + "da55e960029542bf97e9b6190471f818": { + "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 + } + }, + "9e4dbbdf7d694ce491c2bad345e437be": { + "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": "" + } + }, + "7fdb7fdcc62c497e86fb36684ce5a1db": { + "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_e6899428acb04fd2b3271a1a6c07c8fe", + "IPY_MODEL_8b5f6d2bd3fd4712a1cd7e3c00a42cf7", + "IPY_MODEL_a15fad0679f54591ab0c7a796da53544" + ], + "layout": "IPY_MODEL_33bedf27e64749689e39943342c9bacf" + } + }, + "e6899428acb04fd2b3271a1a6c07c8fe": { + "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_03c7da3ceb214befb2419f3bb41d12c7", + "placeholder": "​", + "style": "IPY_MODEL_37eaf99eac1a4424821bb674fb9fdd9f", + "value": "  ...de/consensus_500k/bert.pt: 100%" + } + }, + "8b5f6d2bd3fd4712a1cd7e3c00a42cf7": { + "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_bd788d18620e4310a768c5ecc80f942b", + "max": 1536001492, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_8bf1b010a1da445d90ca090769ff70d7", + "value": 1536001492 + } + }, + "a15fad0679f54591ab0c7a796da53544": { + "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_56d38e95935744a59eafe524d4ad2b31", + "placeholder": "​", + "style": "IPY_MODEL_f3b28548337f4908b469735773ef341d", + "value": " 1.54GB / 1.54GB            " + } + }, + "33bedf27e64749689e39943342c9bacf": { + "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 + } + }, + "03c7da3ceb214befb2419f3bb41d12c7": { + "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 + } + }, + "37eaf99eac1a4424821bb674fb9fdd9f": { + "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": "" + } + }, + "bd788d18620e4310a768c5ecc80f942b": { + "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 + } + }, + "8bf1b010a1da445d90ca090769ff70d7": { + "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": "" + } + }, + "56d38e95935744a59eafe524d4ad2b31": { + "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 + } + }, + "f3b28548337f4908b469735773ef341d": { + "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": "" + } + }, + "a4b3c3215e5d45bcbd6ab93d54264695": { + "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_0d2a88f02744476586ec8844867841c7", + "IPY_MODEL_9b78b5350eff443f936606c5cd447b81", + "IPY_MODEL_80be16cfb01d49dba2fc9b1cd8f46a81" + ], + "layout": "IPY_MODEL_306fe1a1ed994c44ab86cb3991d7692d" + } + }, + "0d2a88f02744476586ec8844867841c7": { + "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_9b1eecbf4ef6497f976daed1aac23d20", + "placeholder": "​", + "style": "IPY_MODEL_8e5eeb51a5c04f119f60758128405426", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "9b78b5350eff443f936606c5cd447b81": { + "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_45aec9e663db4c99a402ce5292ca8587", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_623ac375257342acb2224f4d540f0ddb", + "value": 1 + } + }, + "80be16cfb01d49dba2fc9b1cd8f46a81": { + "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_e150efa1f71443cd86716057c82c1b85", + "placeholder": "​", + "style": "IPY_MODEL_04bdecdb857345d1aa8e7fe072e58e7a", + "value": " 1.54GB / 1.54GB, 9.90MB/s  " + } + }, + "306fe1a1ed994c44ab86cb3991d7692d": { + "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 + } + }, + "9b1eecbf4ef6497f976daed1aac23d20": { + "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 + } + }, + "8e5eeb51a5c04f119f60758128405426": { + "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": "" + } + }, + "45aec9e663db4c99a402ce5292ca8587": { + "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": "20px" + } + }, + "623ac375257342acb2224f4d540f0ddb": { + "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": "" + } + }, + "e150efa1f71443cd86716057c82c1b85": { + "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 + } + }, + "04bdecdb857345d1aa8e7fe072e58e7a": { + "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": "" + } + }, + "d39027271be948c4ac296a71d3562170": { + "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_d88e715865f848ee8171f00e1980a99e", + "IPY_MODEL_afed383eb0e740e1a4bca5fd2fd52bb0", + "IPY_MODEL_962f8493ad9c4bb0a59b6f9cd25d03cc" + ], + "layout": "IPY_MODEL_4b0b4345ad9c4c098cbbce36189c1153" + } + }, + "d88e715865f848ee8171f00e1980a99e": { + "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_84b328f84fd242bc9487a17aa3cf1b0f", + "placeholder": "​", + "style": "IPY_MODEL_0323fa19808b42f9880d3c6e44c30b13", + "value": "New Data Upload               : 100%" + } + }, + "afed383eb0e740e1a4bca5fd2fd52bb0": { + "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_3c931b55244c4027bc6e18fe86511ec8", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_f3f356ad4085424584eefca058a79861", + "value": 1 + } + }, + "962f8493ad9c4bb0a59b6f9cd25d03cc": { + "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_a89bdca1241846e3a05eae708fefbc81", + "placeholder": "​", + "style": "IPY_MODEL_6d6a8ba5f6af4d379f4838e327826e4f", + "value": " 1.54GB / 1.54GB, 9.90MB/s  " + } + }, + "4b0b4345ad9c4c098cbbce36189c1153": { + "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 + } + }, + "84b328f84fd242bc9487a17aa3cf1b0f": { + "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 + } + }, + "0323fa19808b42f9880d3c6e44c30b13": { + "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": "" + } + }, + "3c931b55244c4027bc6e18fe86511ec8": { + "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": "20px" + } + }, + "f3f356ad4085424584eefca058a79861": { + "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": "" + } + }, + "a89bdca1241846e3a05eae708fefbc81": { + "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 + } + }, + "6d6a8ba5f6af4d379f4838e327826e4f": { + "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": "" + } + }, + "459d7be6b5644d12a407e8960209a4d5": { + "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_9b7530dd208249adb88db7b3ae05ba18", + "IPY_MODEL_9d6061b361a14d28a50bf5fc8eb28c23", + "IPY_MODEL_cfa4a32985ff4be98565f980ef2eab18" + ], + "layout": "IPY_MODEL_cc2b2c6a6341499fbf82c949bc275cbe" + } + }, + "9b7530dd208249adb88db7b3ae05ba18": { + "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_d5a58ce74771424780fe9b6fcd8538a9", + "placeholder": "​", + "style": "IPY_MODEL_0fa3ebd0d81f4444970a9ddfd278e3cd", + "value": "  .../consensus_500k/modern.pt: 100%" + } + }, + "9d6061b361a14d28a50bf5fc8eb28c23": { + "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_3fd9968c71274aa581390b1fed4d66f9", + "max": 1536001506, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_644cfc46d21140e1a9e1f96f418e5463", + "value": 1536001506 + } + }, + "cfa4a32985ff4be98565f980ef2eab18": { + "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_e67bd46190674559aca94e0321f7e4ae", + "placeholder": "​", + "style": "IPY_MODEL_6d37220964ac4d3c9f9429ffbd186377", + "value": " 1.54GB / 1.54GB            " + } + }, + "cc2b2c6a6341499fbf82c949bc275cbe": { + "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 + } + }, + "d5a58ce74771424780fe9b6fcd8538a9": { + "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 + } + }, + "0fa3ebd0d81f4444970a9ddfd278e3cd": { + "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": "" + } + }, + "3fd9968c71274aa581390b1fed4d66f9": { + "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 + } + }, + "644cfc46d21140e1a9e1f96f418e5463": { + "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": "" + } + }, + "e67bd46190674559aca94e0321f7e4ae": { + "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 + } + }, + "6d37220964ac4d3c9f9429ffbd186377": { + "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": "" + } + }, + "3aabe6114a6442c2bde35b5ae6ca5e58": { + "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_6c5b50a3caa1424ebbdd373f2dab19a4", + "IPY_MODEL_fe0167cd18664b85ab6b0b10353ad6cd", + "IPY_MODEL_aa1c6b16ec0145e0a56ddf970ef90246" + ], + "layout": "IPY_MODEL_ade78fd5481f4661bba0f50f15fcaaa5" + } + }, + "6c5b50a3caa1424ebbdd373f2dab19a4": { + "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_488a0b550fa44a32a2b3c1cb0d137a77", + "placeholder": "​", + "style": "IPY_MODEL_71de7c2dce1b41af91ff969ba9e60afe", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "fe0167cd18664b85ab6b0b10353ad6cd": { + "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_e799c7d3888641f4a803b00e46c63f3b", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_f0ef1e54e7fc487592b15533a8733a73", + "value": 1 + } + }, + "aa1c6b16ec0145e0a56ddf970ef90246": { + "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_133ac47e177a4bb1ad94a5c5c81b9708", + "placeholder": "​", + "style": "IPY_MODEL_c635d8b0865f4a408ec1bf312d29e8c6", + "value": " 1.54GB / 1.54GB, 7.87MB/s  " + } + }, + "ade78fd5481f4661bba0f50f15fcaaa5": { + "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 + } + }, + "488a0b550fa44a32a2b3c1cb0d137a77": { + "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 + } + }, + "71de7c2dce1b41af91ff969ba9e60afe": { + "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": "" + } + }, + "e799c7d3888641f4a803b00e46c63f3b": { + "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": "20px" + } + }, + "f0ef1e54e7fc487592b15533a8733a73": { + "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": "" + } + }, + "133ac47e177a4bb1ad94a5c5c81b9708": { + "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 + } + }, + "c635d8b0865f4a408ec1bf312d29e8c6": { + "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": "" + } + }, + "b5992e52d30e42018359dd14a88de423": { + "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_72ae7f43086a4f77b0da51ab87ffddb9", + "IPY_MODEL_f3b16ac9b00d49378c2f0b6f4583d53a", + "IPY_MODEL_fad4f7d6a9af4e5e9b31c786b211a512" + ], + "layout": "IPY_MODEL_5d9d457eb7804ae79da4241af0116884" + } + }, + "72ae7f43086a4f77b0da51ab87ffddb9": { + "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_e2b774bddb22445d904b1b6b8403fa27", + "placeholder": "​", + "style": "IPY_MODEL_4e5dd175914445709907787c0f62a0c9", + "value": "New Data Upload               : 100%" + } + }, + "f3b16ac9b00d49378c2f0b6f4583d53a": { + "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_547843a2d89d4151af1a8da52c927dc3", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_f00dcfac18d343f28dedc375f65bcf0f", + "value": 1 + } + }, + "fad4f7d6a9af4e5e9b31c786b211a512": { + "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_e06f304b6495429e937afae8875dacc0", + "placeholder": "​", + "style": "IPY_MODEL_d30624f8745e4996971c52f42bd1e433", + "value": " 1.54GB / 1.54GB, 7.87MB/s  " + } + }, + "5d9d457eb7804ae79da4241af0116884": { + "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 + } + }, + "e2b774bddb22445d904b1b6b8403fa27": { + "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 + } + }, + "4e5dd175914445709907787c0f62a0c9": { + "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": "" + } + }, + "547843a2d89d4151af1a8da52c927dc3": { + "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": "20px" + } + }, + "f00dcfac18d343f28dedc375f65bcf0f": { + "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": "" + } + }, + "e06f304b6495429e937afae8875dacc0": { + "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 + } + }, + "d30624f8745e4996971c52f42bd1e433": { + "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": "" + } + }, + "b105233668c6490b9e8bb1cd2deec55f": { + "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_92284c99f03b4de6b7cb93b87651f71e", + "IPY_MODEL_cf16652f4a0b441685515f512df7bd05", + "IPY_MODEL_4f97a0d0185e4a47981944df05ea0722" + ], + "layout": "IPY_MODEL_e9fdc759f3174970ba20d86ea713890c" + } + }, + "92284c99f03b4de6b7cb93b87651f71e": { + "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_db9bec5b58ab4189954f73ec45c0234d", + "placeholder": "​", + "style": "IPY_MODEL_7c0f34b3d6fa41dfbe25ecb354000243", + "value": "  ...consensus_500k/roberta.pt: 100%" + } + }, + "cf16652f4a0b441685515f512df7bd05": { + "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_a80d180db47b404d82167e86a4bab129", + "max": 1536001577, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ddbd666dfed3416abfc8f0d26be9dadc", + "value": 1536001577 + } + }, + "4f97a0d0185e4a47981944df05ea0722": { + "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_bbe45d6dbd034ba391648511d4b8761e", + "placeholder": "​", + "style": "IPY_MODEL_244d94f8d35840038411f865f5b3f48f", + "value": " 1.54GB / 1.54GB            " + } + }, + "e9fdc759f3174970ba20d86ea713890c": { + "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 + } + }, + "db9bec5b58ab4189954f73ec45c0234d": { + "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 + } + }, + "7c0f34b3d6fa41dfbe25ecb354000243": { + "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": "" + } + }, + "a80d180db47b404d82167e86a4bab129": { + "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 + } + }, + "ddbd666dfed3416abfc8f0d26be9dadc": { + "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": "" + } + }, + "bbe45d6dbd034ba391648511d4b8761e": { + "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 + } + }, + "244d94f8d35840038411f865f5b3f48f": { + "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": "" + } + }, + "d4ab33695163401090a9364f9701a670": { + "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_97cf8215706149518fc00ba3f720bbfe", + "IPY_MODEL_052dd3ef5d074e3da44fedc407ac8926", + "IPY_MODEL_1bfdbf63cfa549ce8d844a9c919cde8a" + ], + "layout": "IPY_MODEL_ced0810573d44581bb8bf4902c84d4cb" + } + }, + "97cf8215706149518fc00ba3f720bbfe": { + "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_9726b22ce55948859094e118438521e7", + "placeholder": "​", + "style": "IPY_MODEL_76a2507027674d0d935ff698064e9fd0", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "052dd3ef5d074e3da44fedc407ac8926": { + "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_dbdb8c81c34b4a18b624053cfd857ab1", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_43980f5d4e15469cab101b804bf8321e", + "value": 1 + } + }, + "1bfdbf63cfa549ce8d844a9c919cde8a": { + "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_118dc7ab862d4c668373c885a0deb53f", + "placeholder": "​", + "style": "IPY_MODEL_18104f6c84324d208c8e6d6476a3ab11", + "value": " 1.54GB / 1.54GB, 16.0MB/s  " + } + }, + "ced0810573d44581bb8bf4902c84d4cb": { + "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 + } + }, + "9726b22ce55948859094e118438521e7": { + "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 + } + }, + "76a2507027674d0d935ff698064e9fd0": { + "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": "" + } + }, + "dbdb8c81c34b4a18b624053cfd857ab1": { + "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": "20px" + } + }, + "43980f5d4e15469cab101b804bf8321e": { + "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": "" + } + }, + "118dc7ab862d4c668373c885a0deb53f": { + "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 + } + }, + "18104f6c84324d208c8e6d6476a3ab11": { + "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": "" + } + }, + "01ae48f5c8eb4961a459010fb763375f": { + "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_9dbaf26407b2429c957a26882371558e", + "IPY_MODEL_0d298d4879a747bfbf62e69f9a3a73de", + "IPY_MODEL_99945a38c13f424cabe8b8edb6d9bc9c" + ], + "layout": "IPY_MODEL_cbdc97ec3cac4aadbf105a52d2c95975" + } + }, + "9dbaf26407b2429c957a26882371558e": { + "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_a3bb128d75404b7f84539f1ccc4253cb", + "placeholder": "​", + "style": "IPY_MODEL_05c165f45f1641a58161ca09ff58b1e2", + "value": "New Data Upload               : 100%" + } + }, + "0d298d4879a747bfbf62e69f9a3a73de": { + "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_0f6619c22c494bb6ba535acc8a56764d", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7258a7ef572c4c7a815d255acf887a22", + "value": 1 + } + }, + "99945a38c13f424cabe8b8edb6d9bc9c": { + "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_b2228d50851a41e8b50739c0bf01a01d", + "placeholder": "​", + "style": "IPY_MODEL_a128d1ab378c4de5951f6efb58d8a9e8", + "value": " 1.54GB / 1.54GB, 16.0MB/s  " + } + }, + "cbdc97ec3cac4aadbf105a52d2c95975": { + "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 + } + }, + "a3bb128d75404b7f84539f1ccc4253cb": { + "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 + } + }, + "05c165f45f1641a58161ca09ff58b1e2": { + "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": "" + } + }, + "0f6619c22c494bb6ba535acc8a56764d": { + "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": "20px" + } + }, + "7258a7ef572c4c7a815d255acf887a22": { + "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": "" + } + }, + "b2228d50851a41e8b50739c0bf01a01d": { + "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 + } + }, + "a128d1ab378c4de5951f6efb58d8a9e8": { + "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": "" + } + }, + "64826e72527f403592c5e1cd02db3cc4": { + "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_96892a8638624211a0ebb6edbac9b291", + "IPY_MODEL_3c3fc5ef68944753b136932dfb295554", + "IPY_MODEL_5b7fb3754b1c4f23bdc92acb0ce3406c" + ], + "layout": "IPY_MODEL_4afc86dddc0543efb4f15f9544ae2ef7" + } + }, + "96892a8638624211a0ebb6edbac9b291": { + "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_2a97d7a2107b488488a03d349c71f6e6", + "placeholder": "​", + "style": "IPY_MODEL_8ece2270cb9c4c3d9591b6340d681519", + "value": "  .../consensus_500k/albert.pt: 100%" + } + }, + "3c3fc5ef68944753b136932dfb295554": { + "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_c3b2417a4cd14d5eb3463a58f7e9a5fb", + "max": 1536001506, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7e7c7a946a934ec3951c39fe2c8b600d", + "value": 1536001506 + } + }, + "5b7fb3754b1c4f23bdc92acb0ce3406c": { + "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_121419d72ac9427a9fe7c29e5c7bf035", + "placeholder": "​", + "style": "IPY_MODEL_96f8af9f08504044bb13345f63f22b3d", + "value": " 1.54GB / 1.54GB            " + } + }, + "4afc86dddc0543efb4f15f9544ae2ef7": { + "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 + } + }, + "2a97d7a2107b488488a03d349c71f6e6": { + "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 + } + }, + "8ece2270cb9c4c3d9591b6340d681519": { + "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": "" + } + }, + "c3b2417a4cd14d5eb3463a58f7e9a5fb": { + "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 + } + }, + "7e7c7a946a934ec3951c39fe2c8b600d": { + "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": "" + } + }, + "121419d72ac9427a9fe7c29e5c7bf035": { + "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 + } + }, + "96f8af9f08504044bb13345f63f22b3d": { + "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": "" + } + }, + "8abef333421b446d89db434f3cf44673": { + "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_4237d6f0583c472bb964a08b9bfd9f86", + "IPY_MODEL_7c8646f25af140888b95fa91abcb9ec8", + "IPY_MODEL_2740216ce3f0424f8077e11e228652fd" + ], + "layout": "IPY_MODEL_66bf0f206ce44a08b7501972c624c7cd" + } + }, + "4237d6f0583c472bb964a08b9bfd9f86": { + "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_781aa0ac2f8c4e11b06fd9a864047333", + "placeholder": "​", + "style": "IPY_MODEL_920f1e10313048ae8db5dd32059b7f55", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "7c8646f25af140888b95fa91abcb9ec8": { + "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_598607abfbfa4afeac842c8f63aabe99", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_4a7e8954f44e4cea85b683a38d377f56", + "value": 1 + } + }, + "2740216ce3f0424f8077e11e228652fd": { + "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_510e6a8a5ea541e79353561a6e026de7", + "placeholder": "​", + "style": "IPY_MODEL_37328c6ca09a4743bf7ea0a0a15aa757", + "value": " 1.54GB / 1.54GB, 16.1MB/s  " + } + }, + "66bf0f206ce44a08b7501972c624c7cd": { + "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 + } + }, + "781aa0ac2f8c4e11b06fd9a864047333": { + "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 + } + }, + "920f1e10313048ae8db5dd32059b7f55": { + "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": "" + } + }, + "598607abfbfa4afeac842c8f63aabe99": { + "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": "20px" + } + }, + "4a7e8954f44e4cea85b683a38d377f56": { + "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": "" + } + }, + "510e6a8a5ea541e79353561a6e026de7": { + "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 + } + }, + "37328c6ca09a4743bf7ea0a0a15aa757": { + "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": "" + } + }, + "452e946cb8b64b549b034f09a477425e": { + "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_0b9ab15129d5494b9c1742a44f2d9229", + "IPY_MODEL_abea04e26d2e401d86e8cbd27b36e57b", + "IPY_MODEL_72355f30c970406598ea0edde4ed5a1d" + ], + "layout": "IPY_MODEL_0b10bd468f5041bfbcb41221a9695e4e" + } + }, + "0b9ab15129d5494b9c1742a44f2d9229": { + "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_1a1846fc50734c538007354f8d1aeecc", + "placeholder": "​", + "style": "IPY_MODEL_12650ad15d5049c5ae2b5edac0f3e098", + "value": "New Data Upload               : 100%" + } + }, + "abea04e26d2e401d86e8cbd27b36e57b": { + "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_687a356cad924ed8b5ec48e3e9dcb02f", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5b888c3082d94c31bf46ef5cc83730ff", + "value": 1 + } + }, + "72355f30c970406598ea0edde4ed5a1d": { + "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_ed65d50863b64a529bc49b33b4592b41", + "placeholder": "​", + "style": "IPY_MODEL_e597ee9e7ffc497ca366143554a23192", + "value": " 1.54GB / 1.54GB, 16.1MB/s  " + } + }, + "0b10bd468f5041bfbcb41221a9695e4e": { + "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 + } + }, + "1a1846fc50734c538007354f8d1aeecc": { + "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 + } + }, + "12650ad15d5049c5ae2b5edac0f3e098": { + "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": "" + } + }, + "687a356cad924ed8b5ec48e3e9dcb02f": { + "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": "20px" + } + }, + "5b888c3082d94c31bf46ef5cc83730ff": { + "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": "" + } + }, + "ed65d50863b64a529bc49b33b4592b41": { + "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 + } + }, + "e597ee9e7ffc497ca366143554a23192": { + "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": "" + } + }, + "3d76035339604e52aaabb287540be49a": { + "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_25ff659b9fe14dd4b5b0c7addce38cdc", + "IPY_MODEL_d4b8b1bdcbd14085a87540f6ec28f470", + "IPY_MODEL_2b6f4a6f1c4f4f24a1ba668e89e2cc00" + ], + "layout": "IPY_MODEL_1356858c469e457fa37f43e675c3e6b0" + } + }, + "25ff659b9fe14dd4b5b0c7addce38cdc": { + "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_11980b18a65748c788cd43401737b53c", + "placeholder": "​", + "style": "IPY_MODEL_a6c15f566fc84f7e90cb3083aef8194b", + "value": "  .../consensus_500k/distil.pt: 100%" + } + }, + "d4b8b1bdcbd14085a87540f6ec28f470": { + "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_42b65b6356b74c959751141b8922fceb", + "max": 1536001506, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5dc1f895e77b42018dcfe3fbffa8567b", + "value": 1536001506 + } + }, + "2b6f4a6f1c4f4f24a1ba668e89e2cc00": { + "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_b48da0a0dcbd40ccb83acf615c754cf7", + "placeholder": "​", + "style": "IPY_MODEL_92ea4373e26d4a6da15ad0f5fa665c45", + "value": " 1.54GB / 1.54GB            " + } + }, + "1356858c469e457fa37f43e675c3e6b0": { + "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 + } + }, + "11980b18a65748c788cd43401737b53c": { + "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 + } + }, + "a6c15f566fc84f7e90cb3083aef8194b": { + "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": "" + } + }, + "42b65b6356b74c959751141b8922fceb": { + "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 + } + }, + "5dc1f895e77b42018dcfe3fbffa8567b": { + "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": "" + } + }, + "b48da0a0dcbd40ccb83acf615c754cf7": { + "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 + } + }, + "92ea4373e26d4a6da15ad0f5fa665c45": { + "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": "" + } + }, + "2a10ca6288854ea4be8ddfc0a112d1ba": { + "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_2991907614494da98e364e7b4b261668", + "IPY_MODEL_3c175404310943b2b975ab724d32c2a7", + "IPY_MODEL_1e063330ec964419a772d37e39adfb26" + ], + "layout": "IPY_MODEL_0f01672fccbf42cf91bb3aca938d1e17" + } + }, + "2991907614494da98e364e7b4b261668": { + "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_5ca32ad1e18143b599789ebad84c0ee9", + "placeholder": "​", + "style": "IPY_MODEL_b776d149237d46ef85acd2c55db891ab", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "3c175404310943b2b975ab724d32c2a7": { + "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_cae01e43ef9548efa678ed58aef7048c", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_f05c0f0910384ec8830c1d436f7a883c", + "value": 1 + } + }, + "1e063330ec964419a772d37e39adfb26": { + "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_964a1b99e204423293abc2233390b5ec", + "placeholder": "​", + "style": "IPY_MODEL_5f3d5b2cb2df45ab92ca01d28e02c04e", + "value": "  232MB /  232MB, 28.3MB/s  " + } + }, + "0f01672fccbf42cf91bb3aca938d1e17": { + "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 + } + }, + "5ca32ad1e18143b599789ebad84c0ee9": { + "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 + } + }, + "b776d149237d46ef85acd2c55db891ab": { + "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": "" + } + }, + "cae01e43ef9548efa678ed58aef7048c": { + "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": "20px" + } + }, + "f05c0f0910384ec8830c1d436f7a883c": { + "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": "" + } + }, + "964a1b99e204423293abc2233390b5ec": { + "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 + } + }, + "5f3d5b2cb2df45ab92ca01d28e02c04e": { + "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": "" + } + }, + "d604e7df44044904816c90d442134d32": { + "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_d3ce02cd5815453d90f60952bb8a8fce", + "IPY_MODEL_85bd6b735db446dca5e92aef832d66f8", + "IPY_MODEL_bdd138b348024ba19efc9d176a90950a" + ], + "layout": "IPY_MODEL_7e83b5c2514142fda892d65abbf3adea" + } + }, + "d3ce02cd5815453d90f60952bb8a8fce": { + "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_cade1456401f42b298a553d4121da01d", + "placeholder": "​", + "style": "IPY_MODEL_7f5678a3a79d483bad005f8ed43a4c76", + "value": "New Data Upload               : 100%" + } + }, + "85bd6b735db446dca5e92aef832d66f8": { + "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_94fea6d51fe1469d92f5283a42639f30", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e9d4c8fbc7464ea490289252403276c9", + "value": 1 + } + }, + "bdd138b348024ba19efc9d176a90950a": { + "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_0211210cc1fc469da30984bd88c9c572", + "placeholder": "​", + "style": "IPY_MODEL_63937d91a06f43e8b6d152b6198def11", + "value": "  232MB /  232MB, 28.3MB/s  " + } + }, + "7e83b5c2514142fda892d65abbf3adea": { + "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 + } + }, + "cade1456401f42b298a553d4121da01d": { + "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 + } + }, + "7f5678a3a79d483bad005f8ed43a4c76": { + "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": "" + } + }, + "94fea6d51fe1469d92f5283a42639f30": { + "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": "20px" + } + }, + "e9d4c8fbc7464ea490289252403276c9": { + "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": "" + } + }, + "0211210cc1fc469da30984bd88c9c572": { + "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 + } + }, + "63937d91a06f43e8b6d152b6198def11": { + "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": "" + } + }, + "fa30381c3e2a48c6a2d2738173667d63": { + "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_6edd557d3a634020999b0c1a47c44069", + "IPY_MODEL_3fdad0148a8c4df187203d1d03dca0ef", + "IPY_MODEL_43df18789778424da8a3fbaf29c33715" + ], + "layout": "IPY_MODEL_65bb38bd08cd449b968dea79f08fd027" + } + }, + "6edd557d3a634020999b0c1a47c44069": { + "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_796a0b14aa4144fe94ca38f09620e5f1", + "placeholder": "​", + "style": "IPY_MODEL_3e477950d41e4d689207fe4de3da3aa8", + "value": "  ...sensus_500k/captions.json: 100%" + } + }, + "3fdad0148a8c4df187203d1d03dca0ef": { + "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_28d93f9b274e428782b6be59665f9984", + "max": 232351545, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_8dcebf5b9b0043b09f3eae87e066798d", + "value": 232351545 + } + }, + "43df18789778424da8a3fbaf29c33715": { + "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_95f6f2f2332d4537860fe746b8a4bd47", + "placeholder": "​", + "style": "IPY_MODEL_4870ff7175c34edbaa2f20fe54d2929a", + "value": "  232MB /  232MB            " + } + }, + "65bb38bd08cd449b968dea79f08fd027": { + "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 + } + }, + "796a0b14aa4144fe94ca38f09620e5f1": { + "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 + } + }, + "3e477950d41e4d689207fe4de3da3aa8": { + "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": "" + } + }, + "28d93f9b274e428782b6be59665f9984": { + "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 + } + }, + "8dcebf5b9b0043b09f3eae87e066798d": { + "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": "" + } + }, + "95f6f2f2332d4537860fe746b8a4bd47": { + "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 + } + }, + "4870ff7175c34edbaa2f20fe54d2929a": { + "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": "" + } + }, + "f46f04d5073945e2add5ac5a82747534": { + "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_36d2eeeb7d6d43e2802991a7ee948df8", + "IPY_MODEL_12d840fff521422eade1ce9792fed91f", + "IPY_MODEL_d31a9a5560d84d5eb94f994682f0454a" + ], + "layout": "IPY_MODEL_b7077c26d0004b82ae2c8b1065804188" + } + }, + "36d2eeeb7d6d43e2802991a7ee948df8": { + "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_5bf5ef0ca2514e5595ed03f2b3d32dd9", + "placeholder": "​", + "style": "IPY_MODEL_1d3d8ca8a3854b0a83db4df3c15f9a6d", + "value": "model.safetensors: 100%" + } + }, + "12d840fff521422eade1ce9792fed91f": { + "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_1d63c2f6c3d641b9b041a0e5142ea891", + "max": 129711020, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_bfcda213611c4943859fc5b8b309a31b", + "value": 129711020 + } + }, + "d31a9a5560d84d5eb94f994682f0454a": { + "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_3ebfa3ed6cd94159bd339766e3a6bb47", + "placeholder": "​", + "style": "IPY_MODEL_329e3afa542d4c788695e60cb541d4a4", + "value": " 130M/130M [00:06<00:00, 36.7MB/s]" + } + }, + "b7077c26d0004b82ae2c8b1065804188": { + "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 + } + }, + "5bf5ef0ca2514e5595ed03f2b3d32dd9": { + "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 + } + }, + "1d3d8ca8a3854b0a83db4df3c15f9a6d": { + "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": "" + } + }, + "1d63c2f6c3d641b9b041a0e5142ea891": { + "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 + } + }, + "bfcda213611c4943859fc5b8b309a31b": { + "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": "" + } + }, + "3ebfa3ed6cd94159bd339766e3a6bb47": { + "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 + } + }, + "329e3afa542d4c788695e60cb541d4a4": { + "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": "" + } + }, + "ea3b041d08e7434498a37af80aed30c3": { + "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_ed5026404e97440e948c9207fc80eaa6", + "IPY_MODEL_8a12c418435c4446ae67db9a0285b27b", + "IPY_MODEL_630ab1aa01bb4b26b430a68ab996a35f" + ], + "layout": "IPY_MODEL_b7b33f7e5d4b445a958fc936350bdd33" + } + }, + "ed5026404e97440e948c9207fc80eaa6": { + "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_288a439304574b978a1a169c934b4f4f", + "placeholder": "​", + "style": "IPY_MODEL_b521d31ddc214bc8a73499fb947615d3", + "value": "Loading weights: 100%" + } + }, + "8a12c418435c4446ae67db9a0285b27b": { + "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_85c078ae16134042996a13d1585ac24f", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_ae5bd77af032451b8d21687f60c4f177", + "value": 112 + } + }, + "630ab1aa01bb4b26b430a68ab996a35f": { + "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_0ea428f6de0d45bd8394be630f8e1d93", + "placeholder": "​", + "style": "IPY_MODEL_a1316b6b0692402e9127f9eec611c6a1", + "value": " 112/112 [00:00<00:00, 3853.86it/s, Materializing param=token_emb.weight]" + } + }, + "b7b33f7e5d4b445a958fc936350bdd33": { + "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 + } + }, + "288a439304574b978a1a169c934b4f4f": { + "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 + } + }, + "b521d31ddc214bc8a73499fb947615d3": { + "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": "" + } + }, + "85c078ae16134042996a13d1585ac24f": { + "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 + } + }, + "ae5bd77af032451b8d21687f60c4f177": { + "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": "" + } + }, + "0ea428f6de0d45bd8394be630f8e1d93": { + "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 + } + }, + "a1316b6b0692402e9127f9eec611c6a1": { + "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": "" + } + }, + "aaa1014322ee46b583d42f9c14926b69": { + "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_5a2fc8ba9e1f40358f2ff2fd5a2417fb", + "IPY_MODEL_699abe3252ac4a0eae550e70da47c4a8", + "IPY_MODEL_d02dd92a6d854d0e8bcb67d313998f4a" + ], + "layout": "IPY_MODEL_a55ca004ee9641f8ade3bfa17b5aaa72" + } + }, + "5a2fc8ba9e1f40358f2ff2fd5a2417fb": { + "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_5203c99f5dce4b928172907f12fa3b66", + "placeholder": "​", + "style": "IPY_MODEL_f4bcfcda00c64a65b6b823d0f8395e1a", + "value": "Loading weights: 100%" + } + }, + "699abe3252ac4a0eae550e70da47c4a8": { + "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_eb4d1532bbec441f877e19215516afa0", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_5fc8bc04e2214dcf81eb2524d533065e", + "value": 112 + } + }, + "d02dd92a6d854d0e8bcb67d313998f4a": { + "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_2b7be4a063034e94aacabc7fad352d3d", + "placeholder": "​", + "style": "IPY_MODEL_5edb67365a5c4bef9c248a99febf42ca", + "value": " 112/112 [00:00<00:00, 3835.07it/s, Materializing param=token_emb.weight]" + } + }, + "a55ca004ee9641f8ade3bfa17b5aaa72": { + "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 + } + }, + "5203c99f5dce4b928172907f12fa3b66": { + "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 + } + }, + "f4bcfcda00c64a65b6b823d0f8395e1a": { + "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": "" + } + }, + "eb4d1532bbec441f877e19215516afa0": { + "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 + } + }, + "5fc8bc04e2214dcf81eb2524d533065e": { + "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": "" + } + }, + "2b7be4a063034e94aacabc7fad352d3d": { + "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 + } + }, + "5edb67365a5c4bef9c248a99febf42ca": { + "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": "" + } + }, + "db10dfddd4e942e1b64e0cda2d3aee64": { + "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_c0396e0a06474078a37a6a84a19d8ce9", + "IPY_MODEL_86ca7df7efda4534b017d17af36fb0e4", + "IPY_MODEL_6a0cecf170354796ad0e64b6a1073c8b" + ], + "layout": "IPY_MODEL_085300d4121a445eaab1864861967ebf" + } + }, + "c0396e0a06474078a37a6a84a19d8ce9": { + "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_aa4dd6dfb06242f19a25050e4eaacf4a", + "placeholder": "​", + "style": "IPY_MODEL_87cd43cc1bc140f99a0f826066a84a4e", + "value": "Loading weights: 100%" + } + }, + "86ca7df7efda4534b017d17af36fb0e4": { + "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_70bfc00d11ef4b2aa607b449abf6b27f", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6c6d9068a112407999016a2e90339756", + "value": 112 + } + }, + "6a0cecf170354796ad0e64b6a1073c8b": { + "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_538cdeaff560479ab2703f3810c4a69d", + "placeholder": "​", + "style": "IPY_MODEL_de747149f30a42c49fb8a11ed518364b", + "value": " 112/112 [00:00<00:00, 3771.22it/s, Materializing param=token_emb.weight]" + } + }, + "085300d4121a445eaab1864861967ebf": { + "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 + } + }, + "aa4dd6dfb06242f19a25050e4eaacf4a": { + "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 + } + }, + "87cd43cc1bc140f99a0f826066a84a4e": { + "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": "" + } + }, + "70bfc00d11ef4b2aa607b449abf6b27f": { + "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 + } + }, + "6c6d9068a112407999016a2e90339756": { + "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": "" + } + }, + "538cdeaff560479ab2703f3810c4a69d": { + "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 + } + }, + "de747149f30a42c49fb8a11ed518364b": { + "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": "" + } + }, + "ef25e8cfba5845d1b4e302267a55af9e": { + "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_a45a6e9a5f294549b6fc2526dabfc4f8", + "IPY_MODEL_d0ddd09569cb402eb9293b6eb509911e", + "IPY_MODEL_d418c749474048318eb434984e396faa" + ], + "layout": "IPY_MODEL_f3cc72fadce64c89808123ce201e9796" + } + }, + "a45a6e9a5f294549b6fc2526dabfc4f8": { + "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_e4a2d1066d044474803ef1b433b4792f", + "placeholder": "​", + "style": "IPY_MODEL_67a56aac4c4d40ad8b109b707347ee45", + "value": "Loading weights: 100%" + } + }, + "d0ddd09569cb402eb9293b6eb509911e": { + "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_eb0326c486d748248eaac16f09399b83", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_9323c9581fcc45eabf61356cdec62562", + "value": 112 + } + }, + "d418c749474048318eb434984e396faa": { + "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_20139b20c7d44c41aa018faf16dc0eee", + "placeholder": "​", + "style": "IPY_MODEL_783f7928d25e4c40ac67d6a173090928", + "value": " 112/112 [00:00<00:00, 3775.62it/s, Materializing param=token_emb.weight]" + } + }, + "f3cc72fadce64c89808123ce201e9796": { + "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 + } + }, + "e4a2d1066d044474803ef1b433b4792f": { + "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 + } + }, + "67a56aac4c4d40ad8b109b707347ee45": { + "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": "" + } + }, + "eb0326c486d748248eaac16f09399b83": { + "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 + } + }, + "9323c9581fcc45eabf61356cdec62562": { + "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": "" + } + }, + "20139b20c7d44c41aa018faf16dc0eee": { + "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 + } + }, + "783f7928d25e4c40ac67d6a173090928": { + "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": "" + } + }, + "b23692bd70a0466a948ed71b0f41a845": { + "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_b8edb67d8fe946b2b2f78a8b7b37bcfb", + "IPY_MODEL_f6c31ab1a52d4cdfa3dc7528cc0fcf3c", + "IPY_MODEL_73c89a1a107c43c58bd14f20defa058f" + ], + "layout": "IPY_MODEL_7a6a4867905f480ca2d43c02f2c198ea" + } + }, + "b8edb67d8fe946b2b2f78a8b7b37bcfb": { + "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_22b1cb0cd8904ecc8f7dcfbca99d27a4", + "placeholder": "​", + "style": "IPY_MODEL_b124a1bc0ccc46e0bd10e03595415358", + "value": "Loading weights: 100%" + } + }, + "f6c31ab1a52d4cdfa3dc7528cc0fcf3c": { + "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_99567e7e1f994314bcf6a190e06a9817", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_6352aa71801b4e97889ddcab35a9e453", + "value": 112 + } + }, + "73c89a1a107c43c58bd14f20defa058f": { + "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_788e0cfd7db4483fb55591c77abb1720", + "placeholder": "​", + "style": "IPY_MODEL_5e814939a89d4afcb80484fac513b5e1", + "value": " 112/112 [00:00<00:00, 3827.04it/s, Materializing param=token_emb.weight]" + } + }, + "7a6a4867905f480ca2d43c02f2c198ea": { + "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 + } + }, + "22b1cb0cd8904ecc8f7dcfbca99d27a4": { + "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 + } + }, + "b124a1bc0ccc46e0bd10e03595415358": { + "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": "" + } + }, + "99567e7e1f994314bcf6a190e06a9817": { + "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 + } + }, + "6352aa71801b4e97889ddcab35a9e453": { + "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": "" + } + }, + "788e0cfd7db4483fb55591c77abb1720": { + "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 + } + }, + "5e814939a89d4afcb80484fac513b5e1": { + "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": "" + } + }, + "27627370b8284c5a862ffd5a3f953352": { + "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_338ff68f8d7445e7bfdb3cbbf8716d86", + "IPY_MODEL_449f0df7844148c59c86f1f6eccf2197", + "IPY_MODEL_a293658c27454a47aadd082610e78d06" + ], + "layout": "IPY_MODEL_1f05373fe992499995c535ec94ab4404" + } + }, + "338ff68f8d7445e7bfdb3cbbf8716d86": { + "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_606a421f64114ed0aed8fc0437f0c171", + "placeholder": "​", + "style": "IPY_MODEL_787e8cc4638e42318b2b93976b04893d", + "value": "Loading weights: 100%" + } + }, + "449f0df7844148c59c86f1f6eccf2197": { + "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_5a76e50e8636432b85c6031fcb303f03", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_c537d231252e4684a0727ec9171e696c", + "value": 112 + } + }, + "a293658c27454a47aadd082610e78d06": { + "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_da8eea806fcb4c9cbd0dbc3d1c6de5ad", + "placeholder": "​", + "style": "IPY_MODEL_18edc062bd4d4c54a496b8c92e2f9d99", + "value": " 112/112 [00:00<00:00, 3915.83it/s, Materializing param=token_emb.weight]" + } + }, + "1f05373fe992499995c535ec94ab4404": { + "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 + } + }, + "606a421f64114ed0aed8fc0437f0c171": { + "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 + } + }, + "787e8cc4638e42318b2b93976b04893d": { + "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": "" + } + }, + "5a76e50e8636432b85c6031fcb303f03": { + "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 + } + }, + "c537d231252e4684a0727ec9171e696c": { + "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": "" + } + }, + "da8eea806fcb4c9cbd0dbc3d1c6de5ad": { + "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 + } + }, + "18edc062bd4d4c54a496b8c92e2f9d99": { + "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": "" + } + }, + "eef3768972ff4b839344f03c681e057d": { + "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_a12a526b205c422dbc9d51765b12c2da", + "IPY_MODEL_df4b6755e47845538d30410ee1ffaef3", + "IPY_MODEL_88c92b8203c8464897ffec2683f1724b" + ], + "layout": "IPY_MODEL_db0cfa00a74749eab32a13a3e81e9426" + } + }, + "a12a526b205c422dbc9d51765b12c2da": { + "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_4831efd2dcb14557af97e6f627a932ec", + "placeholder": "​", + "style": "IPY_MODEL_08b0d9679dc24fa2a86a874c60b67d52", + "value": "Loading weights: 100%" + } + }, + "df4b6755e47845538d30410ee1ffaef3": { + "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_99a3d94fb5dc457da2fa3968e51f2345", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_15a95ffece5e419ab00699f951a71c03", + "value": 112 + } + }, + "88c92b8203c8464897ffec2683f1724b": { + "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_741ee9ffa3084edea972f8f09a877062", + "placeholder": "​", + "style": "IPY_MODEL_d1a3d362dbf9426b9b2b8bb95b0abf52", + "value": " 112/112 [00:00<00:00, 3886.73it/s, Materializing param=token_emb.weight]" + } + }, + "db0cfa00a74749eab32a13a3e81e9426": { + "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 + } + }, + "4831efd2dcb14557af97e6f627a932ec": { + "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 + } + }, + "08b0d9679dc24fa2a86a874c60b67d52": { + "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": "" + } + }, + "99a3d94fb5dc457da2fa3968e51f2345": { + "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 + } + }, + "15a95ffece5e419ab00699f951a71c03": { + "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": "" + } + }, + "741ee9ffa3084edea972f8f09a877062": { + "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 + } + }, + "d1a3d362dbf9426b9b2b8bb95b0abf52": { + "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": "" + } + }, + "d677f32ed3144b6c89f299e88bdd91db": { + "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_89d9aa5b15c443798086d9b42ce6031e", + "IPY_MODEL_a113c72f2b2c403c991c205293f7baa3", + "IPY_MODEL_b8e7f333003c453eb9d4c85d726f56f5" + ], + "layout": "IPY_MODEL_85f4a11c50cc42c9a9e0c9012d72ac04" + } + }, + "89d9aa5b15c443798086d9b42ce6031e": { + "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_0cd8990f98434e1c9f1afdcf6f54a4e9", + "placeholder": "​", + "style": "IPY_MODEL_405998fe651449adaaff88784c1f6c35", + "value": "Loading weights: 100%" + } + }, + "a113c72f2b2c403c991c205293f7baa3": { + "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_1c6231d900844c57b2a0e41d3491d824", + "max": 112, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_a8d419d176c645c4816b9297f15add8f", + "value": 112 + } + }, + "b8e7f333003c453eb9d4c85d726f56f5": { + "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_db04b5a4ac844ee98560ec1ad10e4fbd", + "placeholder": "​", + "style": "IPY_MODEL_9ef880c5e59e4bc8b525b782de253c83", + "value": " 112/112 [00:00<00:00, 3856.51it/s, Materializing param=token_emb.weight]" + } + }, + "85f4a11c50cc42c9a9e0c9012d72ac04": { + "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 + } + }, + "0cd8990f98434e1c9f1afdcf6f54a4e9": { + "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 + } + }, + "405998fe651449adaaff88784c1f6c35": { + "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": "" + } + }, + "1c6231d900844c57b2a0e41d3491d824": { + "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 + } + }, + "a8d419d176c645c4816b9297f15add8f": { + "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": "" + } + }, + "db04b5a4ac844ee98560ec1ad10e4fbd": { + "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 + } + }, + "9ef880c5e59e4bc8b525b782de253c83": { + "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": "" + } + }, + "6a7ca407bed14f36a2f50adced97cfb1": { + "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_043b9f37e147415794ce288ef4ce023b", + "IPY_MODEL_b49cb4fd91514fc6862266e4485dc144", + "IPY_MODEL_6db4e323c2744f99a3850e05c6f52333" + ], + "layout": "IPY_MODEL_a9bd887698ec426c9f0726b79c1214ed" + } + }, + "043b9f37e147415794ce288ef4ce023b": { + "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_28a94a5e2338447caf2ad0c61b977047", + "placeholder": "​", + "style": "IPY_MODEL_562753e117c24e70980bb49556c6d0f2", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "b49cb4fd91514fc6862266e4485dc144": { + "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_bf7130938e2a439b93ae0213e8a0823e", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_b4a41f7f44df427e8c93da8f3efcd397", + "value": 1 + } + }, + "6db4e323c2744f99a3850e05c6f52333": { + "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_0b5121fccdb748b2af32e1a13d8b5aaf", + "placeholder": "​", + "style": "IPY_MODEL_4bc6392ff3d847469ec8c101a274d649", + "value": "  130MB /  130MB, 24.9MB/s  " + } + }, + "a9bd887698ec426c9f0726b79c1214ed": { + "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 + } + }, + "28a94a5e2338447caf2ad0c61b977047": { + "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 + } + }, + "562753e117c24e70980bb49556c6d0f2": { + "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": "" + } + }, + "bf7130938e2a439b93ae0213e8a0823e": { + "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": "20px" + } + }, + "b4a41f7f44df427e8c93da8f3efcd397": { + "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": "" + } + }, + "0b5121fccdb748b2af32e1a13d8b5aaf": { + "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 + } + }, + "4bc6392ff3d847469ec8c101a274d649": { + "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": "" + } + }, + "297ead61ac3346b095e955661c6d793f": { + "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_f7885c7fb502465b996d311fafa3d3cf", + "IPY_MODEL_98675136806f4ff9ba3a862cb25c43ac", + "IPY_MODEL_923a6030739e4ec2aa907f7f6117825f" + ], + "layout": "IPY_MODEL_669117b2232b4470aacffc0fe8083b06" + } + }, + "f7885c7fb502465b996d311fafa3d3cf": { + "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_f070feedd43f45419d830504f3cc6225", + "placeholder": "​", + "style": "IPY_MODEL_d4e81fa7d7d847a19f55f58c34ada30a", + "value": "New Data Upload               : 100%" + } + }, + "98675136806f4ff9ba3a862cb25c43ac": { + "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_76af34506e6c4005b18b3362bbbcf602", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_51861bb2deaa4ca5a894ca37f4e0cce7", + "value": 1 + } + }, + "923a6030739e4ec2aa907f7f6117825f": { + "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_349df077f014467a9014493cda92608d", + "placeholder": "​", + "style": "IPY_MODEL_5c2514217db14f21bb0811ba1ccdf133", + "value": "  129MB /  129MB, 24.8MB/s  " + } + }, + "669117b2232b4470aacffc0fe8083b06": { + "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 + } + }, + "f070feedd43f45419d830504f3cc6225": { + "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 + } + }, + "d4e81fa7d7d847a19f55f58c34ada30a": { + "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": "" + } + }, + "76af34506e6c4005b18b3362bbbcf602": { + "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": "20px" + } + }, + "51861bb2deaa4ca5a894ca37f4e0cce7": { + "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": "" + } + }, + "349df077f014467a9014493cda92608d": { + "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 + } + }, + "5c2514217db14f21bb0811ba1ccdf133": { + "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": "" + } + }, + "23f63ca015404e37bbeff079a0223836": { + "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_a9bbf758389141288b82beb3d1715b1f", + "IPY_MODEL_79ba92bd69d647dba0ee5b3c4fbba100", + "IPY_MODEL_d208af773d8e422fa1ee43748a5d6ca8" + ], + "layout": "IPY_MODEL_c5224ca884a046e7a99a86e6ebc3270a" + } + }, + "a9bbf758389141288b82beb3d1715b1f": { + "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_f2328f11b7cf48798a7c236ae47b2df3", + "placeholder": "​", + "style": "IPY_MODEL_1eb1ad82f20e43f28b75809d54363f0f", + "value": "  /tmp/model.safetensors      : 100%" + } + }, + "79ba92bd69d647dba0ee5b3c4fbba100": { + "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_d1e223c41d424daab3945217f0cd66f9", + "max": 129711020, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_8388fdf7ace342d2aabd8a78a6e99a5d", + "value": 129711020 + } + }, + "d208af773d8e422fa1ee43748a5d6ca8": { + "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_4c1599db89594d7f9ada0e910540e284", + "placeholder": "​", + "style": "IPY_MODEL_1002dd5725b44d4fa9fcf2bb11a63cba", + "value": "  130MB /  130MB            " + } + }, + "c5224ca884a046e7a99a86e6ebc3270a": { + "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 + } + }, + "f2328f11b7cf48798a7c236ae47b2df3": { + "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 + } + }, + "1eb1ad82f20e43f28b75809d54363f0f": { + "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": "" + } + }, + "d1e223c41d424daab3945217f0cd66f9": { + "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 + } + }, + "8388fdf7ace342d2aabd8a78a6e99a5d": { + "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": "" + } + }, + "4c1599db89594d7f9ada0e910540e284": { + "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 + } + }, + "1002dd5725b44d4fa9fcf2bb11a63cba": { + "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": "" + } + }, + "c74c91ea35c840459ed31e6702ba61a4": { + "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_cbe8abf90acc4139ae97a18c52124fec", + "IPY_MODEL_1e10607ea8ab47039a0b2d00517b846f", + "IPY_MODEL_37439a236207473980fd597c5a2c580a" + ], + "layout": "IPY_MODEL_34af6e2c454243a08869979c2bede6fc" + } + }, + "cbe8abf90acc4139ae97a18c52124fec": { + "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_880d5256330a46f696c895a1e999452d", + "placeholder": "​", + "style": "IPY_MODEL_59dcc49ec539497883e8037f0f103211", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "1e10607ea8ab47039a0b2d00517b846f": { + "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_6fa3d96ba7144af398b977131a4e8f33", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_4c3d0e18f2aa4609a962fb4236f848d2", + "value": 1 + } + }, + "37439a236207473980fd597c5a2c580a": { + "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_68c235c748fa40da890e05755833f5a7", + "placeholder": "​", + "style": "IPY_MODEL_c8282090a8234c5c885baecdb6a4139d", + "value": "  130MB /  130MB, 36.7MB/s  " + } + }, + "34af6e2c454243a08869979c2bede6fc": { + "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 + } + }, + "880d5256330a46f696c895a1e999452d": { + "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 + } + }, + "59dcc49ec539497883e8037f0f103211": { + "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": "" + } + }, + "6fa3d96ba7144af398b977131a4e8f33": { + "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": "20px" + } + }, + "4c3d0e18f2aa4609a962fb4236f848d2": { + "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": "" + } + }, + "68c235c748fa40da890e05755833f5a7": { + "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 + } + }, + "c8282090a8234c5c885baecdb6a4139d": { + "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": "" + } + }, + "f4e1a94792c14263a8bfb2c8c72fd70b": { + "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_d1053ee0502b45c2be53200cdb7d6832", + "IPY_MODEL_1a2e4471cec242dd9cba52d6e5a4a96c", + "IPY_MODEL_60d4311784fa440faab8af89af326a25" + ], + "layout": "IPY_MODEL_43f51c9698c04e26bea21284493b6fd8" + } + }, + "d1053ee0502b45c2be53200cdb7d6832": { + "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_88f60f8e6de84f30a6e3a52c5236ddab", + "placeholder": "​", + "style": "IPY_MODEL_6345f7d117ab4d5091fd710ef1f1bec3", + "value": "New Data Upload               : 100%" + } + }, + "1a2e4471cec242dd9cba52d6e5a4a96c": { + "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_99d459ae01df4eee9a139aef9ef5c75b", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_16378ab809a040848bc93688d6df116f", + "value": 1 + } + }, + "60d4311784fa440faab8af89af326a25": { + "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_def85fa4fbe94cbf865d585afa6e78b6", + "placeholder": "​", + "style": "IPY_MODEL_8f6aff29d5ed4a34917f67a4ac5f9fd8", + "value": " 5.34MB / 5.34MB, 3.81MB/s  " + } + }, + "43f51c9698c04e26bea21284493b6fd8": { + "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 + } + }, + "88f60f8e6de84f30a6e3a52c5236ddab": { + "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 + } + }, + "6345f7d117ab4d5091fd710ef1f1bec3": { + "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": "" + } + }, + "99d459ae01df4eee9a139aef9ef5c75b": { + "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": "20px" + } + }, + "16378ab809a040848bc93688d6df116f": { + "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": "" + } + }, + "def85fa4fbe94cbf865d585afa6e78b6": { + "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 + } + }, + "8f6aff29d5ed4a34917f67a4ac5f9fd8": { + "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": "" + } + }, + "c1f908de263f4289bc5e47bc2c2d1c99": { + "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_13909be31fe94bc586eebaefe1085762", + "IPY_MODEL_622faec06d4e4b5fbbf98dbf1c3b1825", + "IPY_MODEL_f62c112453c14983bf264e2d074a21ee" + ], + "layout": "IPY_MODEL_98dc253bc2694ce5890fae769962428b" + } + }, + "13909be31fe94bc586eebaefe1085762": { + "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_ac4437a9cfa64a81bb6facaadc54ea66", + "placeholder": "​", + "style": "IPY_MODEL_bda406c344974476a3c6187f1b4db7a0", + "value": "  cotrain_best.pt             : 100%" + } + }, + "622faec06d4e4b5fbbf98dbf1c3b1825": { + "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_f5d44be362dc47389738eb8c0576e9df", + "max": 129740397, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_dcf98d4aae334ba99cd2f293a60f8b8c", + "value": 129740397 + } + }, + "f62c112453c14983bf264e2d074a21ee": { + "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_c98d1f2961b24a77b866c798516bdcac", + "placeholder": "​", + "style": "IPY_MODEL_ba070298a24043bfae6d0f476370bdab", + "value": "  130MB /  130MB            " + } + }, + "98dc253bc2694ce5890fae769962428b": { + "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 + } + }, + "ac4437a9cfa64a81bb6facaadc54ea66": { + "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 + } + }, + "bda406c344974476a3c6187f1b4db7a0": { + "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": "" + } + }, + "f5d44be362dc47389738eb8c0576e9df": { + "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 + } + }, + "dcf98d4aae334ba99cd2f293a60f8b8c": { + "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": "" + } + }, + "c98d1f2961b24a77b866c798516bdcac": { + "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 + } + }, + "ba070298a24043bfae6d0f476370bdab": { + "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": "" + } + }, + "527d741b6dad45fca647c29cd0f0d51e": { + "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_737153f743a34357a80fa00477c62ea8", + "IPY_MODEL_6276af2526ec42b7b92986c2c50a7cc0", + "IPY_MODEL_df44d06b4ca34e799a7f3260dd556e0d" + ], + "layout": "IPY_MODEL_1bcc6bf383294fffaddf8d6c0a8c6b3f" + } + }, + "737153f743a34357a80fa00477c62ea8": { + "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_41a02a3c6d22426596ad49dbc6a43bf9", + "placeholder": "​", + "style": "IPY_MODEL_34170e57078d47228bbee8489a153069", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "6276af2526ec42b7b92986c2c50a7cc0": { + "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_62a3861e3f6d4987ad3ab7d0cde767e4", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_e4b930bee8ee4daea154e1dfef773edc", + "value": 1 + } + }, + "df44d06b4ca34e799a7f3260dd556e0d": { + "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_5481cc1f34624726ae71e2c1150d9f27", + "placeholder": "​", + "style": "IPY_MODEL_e77960ab9e2e4dd3b46658a67cc27c80", + "value": "  130MB /  130MB, 27.0MB/s  " + } + }, + "1bcc6bf383294fffaddf8d6c0a8c6b3f": { + "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 + } + }, + "41a02a3c6d22426596ad49dbc6a43bf9": { + "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 + } + }, + "34170e57078d47228bbee8489a153069": { + "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": "" + } + }, + "62a3861e3f6d4987ad3ab7d0cde767e4": { + "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": "20px" + } + }, + "e4b930bee8ee4daea154e1dfef773edc": { + "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": "" + } + }, + "5481cc1f34624726ae71e2c1150d9f27": { + "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 + } + }, + "e77960ab9e2e4dd3b46658a67cc27c80": { + "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": "" + } + }, + "fff6def384484ff483e3f6dc8f233fcb": { + "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_51443b657d0741ff898c78a76a723d09", + "IPY_MODEL_d93efc0c12604e969be62a12686f7557", + "IPY_MODEL_054c2b70cebd466f9715c12ac4688356" + ], + "layout": "IPY_MODEL_f34252d66df9427d8bbe8afdec7a92ca" + } + }, + "51443b657d0741ff898c78a76a723d09": { + "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_8558b4cea4cf4b6aa57b75e9108d4bb2", + "placeholder": "​", + "style": "IPY_MODEL_ca1c3b78d9104efe878545a8098dc982", + "value": "New Data Upload               : 100%" + } + }, + "d93efc0c12604e969be62a12686f7557": { + "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_ff19e48299d34a108a00ecef63a68a85", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7b6328b4b0ee43359e071329951ab267", + "value": 1 + } + }, + "054c2b70cebd466f9715c12ac4688356": { + "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_afe9e7959c484552b441b735dfdc1b49", + "placeholder": "​", + "style": "IPY_MODEL_2571a6938bd4424ba5a5fee9e268575f", + "value": "  129MB /  129MB, 26.9MB/s  " + } + }, + "f34252d66df9427d8bbe8afdec7a92ca": { + "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 + } + }, + "8558b4cea4cf4b6aa57b75e9108d4bb2": { + "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 + } + }, + "ca1c3b78d9104efe878545a8098dc982": { + "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": "" + } + }, + "ff19e48299d34a108a00ecef63a68a85": { + "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": "20px" + } + }, + "7b6328b4b0ee43359e071329951ab267": { + "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": "" + } + }, + "afe9e7959c484552b441b735dfdc1b49": { + "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 + } + }, + "2571a6938bd4424ba5a5fee9e268575f": { + "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": "" + } + }, + "7e074525092d4290a09967b2b2abcfb5": { + "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_ad7dec632eae499f865dc8c051f133c8", + "IPY_MODEL_3707b8f64cd74588901207d1159fbf0c", + "IPY_MODEL_064824f6b6fc4478b43209ca4997341b" + ], + "layout": "IPY_MODEL_757fdfd851be4bb6a6ea2c161d96f8e8" + } + }, + "ad7dec632eae499f865dc8c051f133c8": { + "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_3cbf2b6ebe20448eacad73b847da784b", + "placeholder": "​", + "style": "IPY_MODEL_d39e32cc041d4e5e8c26f38d9a82dc14", + "value": "  cotrain_final.pt            : 100%" + } + }, + "3707b8f64cd74588901207d1159fbf0c": { + "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_28d143b1705e4e089a35320d6adad602", + "max": 129740515, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_1add89288ec24124883af4e84b355f97", + "value": 129740515 + } + }, + "064824f6b6fc4478b43209ca4997341b": { + "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_83c9d8c9ac5b4d478e445cd6fc647262", + "placeholder": "​", + "style": "IPY_MODEL_94a2d86779e7425bb94805cd11b77671", + "value": "  130MB /  130MB            " + } + }, + "757fdfd851be4bb6a6ea2c161d96f8e8": { + "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 + } + }, + "3cbf2b6ebe20448eacad73b847da784b": { + "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 + } + }, + "d39e32cc041d4e5e8c26f38d9a82dc14": { + "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": "" + } + }, + "28d143b1705e4e089a35320d6adad602": { + "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 + } + }, + "1add89288ec24124883af4e84b355f97": { + "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": "" + } + }, + "83c9d8c9ac5b4d478e445cd6fc647262": { + "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 + } + }, + "94a2d86779e7425bb94805cd11b77671": { + "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": "" + } + }, + "56c086ce70b94a658e650089a327ba23": { + "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_f9ef6cea18ad4b258dcba578f0483886", + "IPY_MODEL_699967cd5d8e48d6979a64c1d6d1e996", + "IPY_MODEL_2f68aa30b3f24b01b4f0d1b42ab6d4ab" + ], + "layout": "IPY_MODEL_d6daf985235049828d51fd4b2d5c81e1" + } + }, + "f9ef6cea18ad4b258dcba578f0483886": { + "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_2131b22b413547aea0fbedb5de878c09", + "placeholder": "​", + "style": "IPY_MODEL_73c3bde3ff67486087769238aeb84075", + "value": "Processing Files (1 / 1)      : 100%" + } + }, + "699967cd5d8e48d6979a64c1d6d1e996": { + "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_45d2fd69b8e34c43ba9ec944869f4777", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7c25478f05274a898435f32475707e34", + "value": 1 + } + }, + "2f68aa30b3f24b01b4f0d1b42ab6d4ab": { + "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_d998b870eead442e96ff30f12257d5dc", + "placeholder": "​", + "style": "IPY_MODEL_112f3f5d5283473ab40dabf47a8ae298", + "value": "  224kB /  224kB,  560kB/s  " + } + }, + "d6daf985235049828d51fd4b2d5c81e1": { + "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 + } + }, + "2131b22b413547aea0fbedb5de878c09": { + "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 + } + }, + "73c3bde3ff67486087769238aeb84075": { + "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": "" + } + }, + "45d2fd69b8e34c43ba9ec944869f4777": { + "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": "20px" + } + }, + "7c25478f05274a898435f32475707e34": { + "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": "" + } + }, + "d998b870eead442e96ff30f12257d5dc": { + "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 + } + }, + "112f3f5d5283473ab40dabf47a8ae298": { + "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": "" + } + }, + "40916ab338ef40d6baefb3d293bbd751": { + "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_f7e25c729a2e4602bfa478039c98b3c5", + "IPY_MODEL_a2277a303291474d86a9d6c51eeb254a", + "IPY_MODEL_e5ffe88aabe74ed9973e620f2d88ec0f" + ], + "layout": "IPY_MODEL_9ec88e223c7a4fa190d678d9597db0a1" + } + }, + "f7e25c729a2e4602bfa478039c98b3c5": { + "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_af64efc892034b70a6b053f3c741b943", + "placeholder": "​", + "style": "IPY_MODEL_dbf69d41b620475e98f441b8f05a7c42", + "value": "New Data Upload               : 100%" + } + }, + "a2277a303291474d86a9d6c51eeb254a": { + "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_8295bb93d162410e9a8f667770da975b", + "max": 1, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_d0cacf37b2fb476e9a4c2b17a615779e", + "value": 1 + } + }, + "e5ffe88aabe74ed9973e620f2d88ec0f": { + "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_56279fa5fa6144cd83f6ac9fa704b7eb", + "placeholder": "​", + "style": "IPY_MODEL_9d67bb98d74c4233b1d4de1c8705093c", + "value": "  224kB /  224kB,  560kB/s  " + } + }, + "9ec88e223c7a4fa190d678d9597db0a1": { + "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 + } + }, + "af64efc892034b70a6b053f3c741b943": { + "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 + } + }, + "dbf69d41b620475e98f441b8f05a7c42": { + "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": "" + } + }, + "8295bb93d162410e9a8f667770da975b": { + "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": "20px" + } + }, + "d0cacf37b2fb476e9a4c2b17a615779e": { + "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": "" + } + }, + "56279fa5fa6144cd83f6ac9fa704b7eb": { + "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 + } + }, + "9d67bb98d74c4233b1d4de1c8705093c": { + "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": "" + } + }, + "fa25828282bf4b979721afcb2657faed": { + "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_b049721738824b3d9ffcf381689367f8", + "IPY_MODEL_7b14bdeff33e481797fd3e538716b4f2", + "IPY_MODEL_4b40034aae194d18b8e2088951e1e0c4" + ], + "layout": "IPY_MODEL_15cf6a3aafa64b4f871b2ffe80169c9b" + } + }, + "b049721738824b3d9ffcf381689367f8": { + "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_35002ec31ac343b4b1c7cd3e0a979e95", + "placeholder": "​", + "style": "IPY_MODEL_a0b8937dec184de1b22bbb75f0cf6ef0", + "value": "  ...433.8fa90824b991.141712.0: 100%" + } + }, + "7b14bdeff33e481797fd3e538716b4f2": { + "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_29b32216f02a460d8d591eb91d119262", + "max": 223999, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_7f3547274091448eb5afe8bb23129680", + "value": 223999 + } + }, + "4b40034aae194d18b8e2088951e1e0c4": { + "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_87cd0c2ff682478691952c3895e11c1a", + "placeholder": "​", + "style": "IPY_MODEL_5bab38cd67cd4b78ba2afbdd2226db4e", + "value": "  224kB /  224kB            " + } + }, + "15cf6a3aafa64b4f871b2ffe80169c9b": { + "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 + } + }, + "35002ec31ac343b4b1c7cd3e0a979e95": { + "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 + } + }, + "a0b8937dec184de1b22bbb75f0cf6ef0": { + "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": "" + } + }, + "29b32216f02a460d8d591eb91d119262": { + "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 + } + }, + "7f3547274091448eb5afe8bb23129680": { + "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": "" + } + }, + "87cd0c2ff682478691952c3895e11c1a": { + "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 + } + }, + "5bab38cd67cd4b78ba2afbdd2226db4e": { + "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": "" + } + } + } + } + }, + "cells": [ + { + "cell_type": "markdown", + "source": [ + "# conv5d example" + ], + "metadata": { + "id": "Yzc_4StXPqN_" + } + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "4GwHQAIkPLvS", + "outputId": "60538d2e-cf95-4f47-b1d6-0245c1b24455" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "============================================================\n", + "COMPOSITIONAL CONVOLUTION: PARTITION ANALYSIS\n", + "============================================================\n", + "\n", + "Compositions of 4: 8 total\n", + "Composition Length Interpretation\n", + "------------------------------------------------------------\n", + " (4,) 1 4-way\n", + " (1, 3) 2 independent → 3-way\n", + " (2, 2) 2 pairwise → pairwise\n", + " (3, 1) 2 3-way → independent\n", + " (1, 1, 2) 3 independent → independent → pairwise\n", + " (1, 2, 1) 3 independent → pairwise → independent\n", + " (2, 1, 1) 3 pairwise → independent → independent\n", + " (1, 1, 1, 1) 4 independent → independent → independent → independent\n", + "\n", + "Compositions of 5: 16 total\n", + "Composition Length Interpretation\n", + "------------------------------------------------------------\n", + " (5,) 1 5-way (full)\n", + " (1, 4) 2 independent → 4-way\n", + " (2, 3) 2 pairwise → 3-way\n", + " (3, 2) 2 3-way → pairwise\n", + " (4, 1) 2 4-way → independent\n", + " (1, 1, 3) 3 independent → independent → 3-way\n", + " (1, 2, 2) 3 independent → pairwise → pairwise\n", + " (1, 3, 1) 3 independent → 3-way → independent\n", + " (2, 1, 2) 3 pairwise → independent → pairwise\n", + " (2, 2, 1) 3 pairwise → pairwise → independent\n", + " (3, 1, 1) 3 3-way → independent → independent\n", + " (1, 1, 1, 2) 4 independent → independent → independent → pairwise\n", + " (1, 1, 2, 1) 4 independent → independent → pairwise → independent\n", + " (1, 2, 1, 1) 4 independent → pairwise → independent → independent\n", + " (2, 1, 1, 1) 4 pairwise → independent → independent → independent\n", + " (1, 1, 1, 1, 1) 5 independent → independent → independent → independent → independent\n", + "\n", + "conv4d: 8 paths (vs 1 opaque operator)\n", + "conv5d: 16 paths (vs 1 opaque operator)\n", + "\n", + "The 16 paths for conv5d enumerate ALL ways to\n", + "traverse a 5-dimensional simplex (pentachoron).\n", + "Each path captures a different structural relationship.\n", + "Together they form a complete basis for 5d geometry.\n", + "\n", + "============================================================\n", + "TESTING: DualAnchorEmbedding with conv4\n", + "============================================================\n", + " Parameters: 12,951,816\n", + " Conv4 paths: 8\n", + " Path 0: (1, 1, 1, 1)\n", + " Path 1: (1, 1, 2)\n", + " Path 2: (1, 2, 1)\n", + " Path 3: (1, 3)\n", + " Path 4: (2, 1, 1)\n", + " Path 5: (2, 2)\n", + " Path 6: (3, 1)\n", + " Path 7: (4,)\n", + "\n", + " Input: anchor_a=(16, 768), anchor_b=(16, 768)\n", + " Output: (16, 768)\n", + " Output norm: 1.0000\n", + "\n", + " Path weights (learned):\n", + " (1, 1, 1, 1) weight=0.1250\n", + " (1, 1, 2) weight=0.1250\n", + " (1, 2, 1) weight=0.1250\n", + " (1, 3) weight=0.1250\n", + " (2, 1, 1) weight=0.1250\n", + " (2, 2) weight=0.1250\n", + " (3, 1) weight=0.1250\n", + " (4,) weight=0.1250\n", + "\n", + "Done.\n" + ] + } + ], + "source": [ + "# ============================================================================\n", + "# EXPERIMENT: Compositional Convolution via Integer Partitions\n", + "#\n", + "# Hypothesis: conv4d can be decomposed into the set of ordered compositions\n", + "# of 4, each capturing a different factorization of 4-dimensional structure.\n", + "# The union of all compositions is a complete, interpretable, well-conditioned\n", + "# alternative to opaque N-dimensional convolution.\n", + "#\n", + "# Applied to: structural differences between two Procrustes-aligned BERTs,\n", + "# accumulated in a geometric memory bank.\n", + "#\n", + "# The conv4 partition set becomes the abstraction to build conv5d (pentachoron).\n", + "#\n", + "# Compositions of 4: [1,1,1,1], [1,1,2], [1,2,1], [2,1,1], [2,2], [3,1], [1,3], [4]\n", + "# Compositions of 5: 16 total paths\n", + "# ============================================================================\n", + "\n", + "import math\n", + "from typing import List, Tuple\n", + "from itertools import product as cartesian\n", + "\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# COMPOSITION ENUMERATION\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def integer_compositions(n: int) -> List[Tuple[int, ...]]:\n", + " \"\"\"\n", + " Generate all ordered compositions of integer n.\n", + " e.g. n=4 → (1,1,1,1), (1,1,2), (1,2,1), (2,1,1), (2,2), (1,3), (3,1), (4)\n", + " \"\"\"\n", + " if n == 0:\n", + " return [()]\n", + " if n == 1:\n", + " return [(1,)]\n", + " result = []\n", + " for first in range(1, n + 1):\n", + " for rest in integer_compositions(n - first):\n", + " result.append((first,) + rest)\n", + " return result\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# COMPOSITIONAL CONV PATH\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class ConvPath(nn.Module):\n", + " \"\"\"\n", + " A single composition path through N-dimensional structure.\n", + "\n", + " Given composition (a, b, c, ...) where a+b+c+...=N:\n", + " - Apply conv_{a}d, then conv_{b}d, then conv_{c}d\n", + " - Each step processes a different factorization of the input dimensions\n", + "\n", + " For embedding space: dimensions are abstract (not spatial),\n", + " so we use grouped linear projections to simulate N-d convolution\n", + " over the embedding's geometric structure.\n", + "\n", + " conv1d over embeddings = project single dim slices independently\n", + " conv2d over embeddings = project pairs of dim slices jointly\n", + " conv_kd over embeddings = project k-dim slices jointly\n", + " \"\"\"\n", + " def __init__(self, composition: Tuple[int, ...], embed_dim: int, hidden_dim: int):\n", + " super().__init__()\n", + " self.composition = composition\n", + " self.total_n = sum(composition)\n", + " self.embed_dim = embed_dim\n", + " self.hidden_dim = hidden_dim\n", + "\n", + " # Each step in the composition gets a projection\n", + " self.steps = nn.ModuleList()\n", + " current_dim = embed_dim\n", + " for k in composition:\n", + " # conv_kd equivalent: project groups of k dimensions jointly\n", + " # Reshape embed into groups, project each group\n", + " self.steps.append(nn.ModuleDict({\n", + " \"proj\": nn.Linear(current_dim, hidden_dim),\n", + " \"group_mix\": nn.Linear(hidden_dim, hidden_dim),\n", + " \"norm\": nn.LayerNorm(hidden_dim),\n", + " \"k\": nn.Module(), # placeholder for k value\n", + " }))\n", + " # Store k as buffer\n", + " self.steps[-1].k_value = k\n", + " current_dim = hidden_dim\n", + "\n", + " self.output_proj = nn.Linear(hidden_dim, embed_dim)\n", + "\n", + " def forward(self, x):\n", + " \"\"\"\n", + " x: (B, embed_dim) — the geometric difference between two anchor views\n", + "\n", + " Each step reshapes into groups of k dimensions, processes jointly,\n", + " then flattens back. This captures k-dimensional correlations at each stage.\n", + " \"\"\"\n", + " B = x.shape[0]\n", + " h = x\n", + "\n", + " for step in self.steps:\n", + " k = step.k_value\n", + "\n", + " # Project to hidden space\n", + " h = step[\"proj\"](h)\n", + " h = F.gelu(h)\n", + "\n", + " # Group mixing: reshape into groups of (hidden_dim // k) × k\n", + " # then mix within groups to capture k-dimensional correlations\n", + " if k > 1 and self.hidden_dim >= k:\n", + " n_groups = self.hidden_dim // k\n", + " remainder = self.hidden_dim % k\n", + " if n_groups > 0 and remainder == 0:\n", + " grouped = h.view(B, n_groups, k)\n", + " # k-dimensional interaction within each group\n", + " grouped = grouped * torch.softmax(grouped, dim=-1)\n", + " h = grouped.view(B, self.hidden_dim)\n", + "\n", + " h = step[\"group_mix\"](h)\n", + " h = F.gelu(h)\n", + " h = step[\"norm\"](h)\n", + "\n", + " return self.output_proj(h)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# COMPOSITIONAL CONV-N MODULE\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class CompositionalConvN(nn.Module):\n", + " \"\"\"\n", + " All compositions of N running in parallel.\n", + "\n", + " Each composition path captures a different factorization of\n", + " N-dimensional structure. The outputs are fused via learned\n", + " attention weighting.\n", + "\n", + " For N=4:\n", + " 8 paths: [1,1,1,1], [1,1,2], [1,2,1], [2,1,1], [2,2], [1,3], [3,1], [4]\n", + "\n", + " For N=5 (pentachoron):\n", + " 16 paths: all compositions of 5\n", + " \"\"\"\n", + " def __init__(self, n: int, embed_dim: int, hidden_dim: int,\n", + " max_paths: int = 32):\n", + " super().__init__()\n", + " self.n = n\n", + " self.compositions = integer_compositions(n)\n", + "\n", + " # Optionally limit paths for very large N\n", + " if len(self.compositions) > max_paths:\n", + " # Keep shortest, longest, and sample middle\n", + " sorted_comps = sorted(self.compositions, key=len)\n", + " keep = set()\n", + " keep.add(sorted_comps[0]) # shortest (N,)\n", + " keep.add(sorted_comps[-1]) # longest (1,1,...,1)\n", + " # Evenly sample the rest\n", + " step = max(len(sorted_comps) // max_paths, 1)\n", + " for i in range(0, len(sorted_comps), step):\n", + " keep.add(sorted_comps[i])\n", + " self.compositions = list(keep)[:max_paths]\n", + "\n", + " self.paths = nn.ModuleList([\n", + " ConvPath(comp, embed_dim, hidden_dim)\n", + " for comp in self.compositions\n", + " ])\n", + "\n", + " # Fusion: learned attention over path outputs\n", + " self.n_paths = len(self.paths)\n", + " self.path_weights = nn.Parameter(torch.ones(self.n_paths) / self.n_paths)\n", + " self.fusion = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim),\n", + " nn.GELU(),\n", + " nn.LayerNorm(embed_dim),\n", + " )\n", + "\n", + " def forward(self, x):\n", + " \"\"\"\n", + " x: (B, embed_dim)\n", + " Returns: (B, embed_dim) — fused output from all composition paths\n", + " \"\"\"\n", + " outputs = torch.stack([path(x) for path in self.paths], dim=1)\n", + " # (B, n_paths, embed_dim)\n", + "\n", + " weights = F.softmax(self.path_weights, dim=0)\n", + " fused = (outputs * weights.view(1, -1, 1)).sum(dim=1)\n", + " return self.fusion(fused)\n", + "\n", + " def path_analysis(self, x):\n", + " \"\"\"Diagnostic: return per-path outputs and weights for analysis.\"\"\"\n", + " outputs = [path(x) for path in self.paths]\n", + " weights = F.softmax(self.path_weights, dim=0)\n", + " return {\n", + " \"compositions\": self.compositions,\n", + " \"weights\": weights.detach().cpu(),\n", + " \"outputs\": [o.detach().cpu() for o in outputs],\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRIC STRUCTURAL MEMORY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class StructuralMemoryBank(nn.Module):\n", + " \"\"\"\n", + " Memory bank that stores compositional conv decompositions of the\n", + " structural differences between two Procrustes-aligned anchor views.\n", + "\n", + " Instead of anchoring to model layers (like the CLIP memory bank),\n", + " this anchors to an embedding spectrum — the full set of compositional\n", + " decompositions of the geometric difference between experts.\n", + "\n", + " Each memory slot stores:\n", + " - The fused compositional output (how A and B differ)\n", + " - Individual path activations (which factorizations are active)\n", + " - Geometric regularity (pentachoron CV on the slot ensemble)\n", + " \"\"\"\n", + " def __init__(self, embed_dim: int, hidden_dim: int, conv_n: int = 4,\n", + " bank_size: int = 64, n_heads: int = 8):\n", + " super().__init__()\n", + " self.embed_dim = embed_dim\n", + " self.bank_size = bank_size\n", + "\n", + " # Compositional conv for structural difference\n", + " self.conv_n = CompositionalConvN(conv_n, embed_dim, hidden_dim)\n", + "\n", + " # Bank storage\n", + " self.register_buffer(\n", + " \"bank_slots\", torch.zeros(bank_size, embed_dim))\n", + " self.register_buffer(\"n_written\", torch.tensor(0, dtype=torch.long))\n", + "\n", + " # Read mechanism: cross-attention over bank\n", + " self.read_attn = nn.MultiheadAttention(\n", + " embed_dim, n_heads, batch_first=True, dropout=0.1)\n", + " self.read_norm = nn.LayerNorm(embed_dim)\n", + " self.read_ffn = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim * 2),\n", + " nn.GELU(),\n", + " nn.Linear(embed_dim * 2, embed_dim))\n", + " self.read_ffn_norm = nn.LayerNorm(embed_dim)\n", + "\n", + " def write(self, anchor_a: torch.Tensor, anchor_b: torch.Tensor):\n", + " \"\"\"\n", + " anchor_a, anchor_b: (B, embed_dim) — Procrustes-aligned embeddings\n", + " Computes structural difference, decomposes via compositional conv,\n", + " writes to bank.\n", + " \"\"\"\n", + " # Geometric structural difference — not just subtraction\n", + " diff = anchor_a - anchor_b\n", + " product = anchor_a * anchor_b\n", + " mean = (anchor_a + anchor_b) / 2\n", + " structural = torch.cat([diff, product, mean], dim=-1)\n", + "\n", + " # Project back to embed_dim for conv processing\n", + " if not hasattr(self, \"struct_proj\"):\n", + " self.struct_proj = nn.Linear(\n", + " self.embed_dim * 3, self.embed_dim\n", + " ).to(anchor_a.device)\n", + " structural = self.struct_proj(structural)\n", + "\n", + " # Decompose through all composition paths\n", + " decomposed = self.conv_n(structural) # (B, embed_dim)\n", + "\n", + " # Write to bank (circular buffer)\n", + " B = decomposed.shape[0]\n", + " for i in range(B):\n", + " idx = self.n_written % self.bank_size\n", + " self.bank_slots[idx] = decomposed[i].detach()\n", + " self.n_written += 1\n", + "\n", + " return decomposed\n", + "\n", + " def read(self, query: torch.Tensor) -> torch.Tensor:\n", + " \"\"\"\n", + " query: (B, seq, embed_dim) — what to condition on bank content\n", + " Returns: (B, seq, embed_dim) — enriched by bank memory\n", + " \"\"\"\n", + " n = min(self.n_written.item(), self.bank_size)\n", + " if n == 0:\n", + " return query\n", + "\n", + " B = query.shape[0]\n", + " bank = self.bank_slots[:n].unsqueeze(0).expand(B, -1, -1)\n", + "\n", + " # Cross-attend: query reads from bank\n", + " residual = query\n", + " q_normed = self.read_norm(query)\n", + " attended, _ = self.read_attn(q_normed, bank, bank)\n", + " query = residual + attended\n", + "\n", + " residual = query\n", + " query = residual + self.read_ffn(self.read_ffn_norm(query))\n", + "\n", + " return query\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXPERIMENT: TWO-BERT STRUCTURAL EMBEDDING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class DualAnchorEmbedding(nn.Module):\n", + " \"\"\"\n", + " Full experiment module:\n", + " 1. Two frozen BERTs produce aligned embeddings\n", + " 2. Compositional conv decomposes their structural difference\n", + " 3. Memory bank accumulates decompositions\n", + " 4. Output embedding = bank-enriched representation\n", + "\n", + " The embedding spectrum is the UNION of all compositional views\n", + " of how two different trained models interpret the same input.\n", + " \"\"\"\n", + " def __init__(self, embed_dim: int = 768, hidden_dim: int = 256,\n", + " conv_n: int = 4, bank_size: int = 64):\n", + " super().__init__()\n", + " self.embed_dim = embed_dim\n", + "\n", + " # Structural memory with compositional conv\n", + " self.memory = StructuralMemoryBank(\n", + " embed_dim, hidden_dim, conv_n, bank_size)\n", + "\n", + " # Learner: produces unified embedding from dual-anchor context\n", + " self.embed_proj = nn.Sequential(\n", + " nn.Linear(embed_dim * 2, embed_dim),\n", + " nn.GELU(),\n", + " nn.LayerNorm(embed_dim))\n", + "\n", + " # Output head\n", + " self.output_head = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim),\n", + " nn.GELU(),\n", + " nn.Linear(embed_dim, embed_dim))\n", + "\n", + " def forward(self, anchor_a: torch.Tensor, anchor_b: torch.Tensor):\n", + " \"\"\"\n", + " anchor_a: (B, embed_dim) — BERT-base aligned\n", + " anchor_b: (B, embed_dim) — ModernBERT aligned\n", + "\n", + " Returns: (B, embed_dim) — unified geometric embedding\n", + " \"\"\"\n", + " # Write structural difference to memory\n", + " structural = self.memory.write(anchor_a, anchor_b)\n", + "\n", + " # Combine both anchors\n", + " combined = self.embed_proj(torch.cat([anchor_a, anchor_b], dim=-1))\n", + "\n", + " # Read from bank (use combined as query)\n", + " enriched = self.memory.read(combined.unsqueeze(1)).squeeze(1)\n", + "\n", + " # Fuse: structural decomposition + bank-enriched combined\n", + " fused = enriched + structural\n", + "\n", + " return F.normalize(self.output_head(fused), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# ANALYSIS TOOLS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def analyze_compositions(n: int):\n", + " \"\"\"Print all compositions of n with their structural interpretation.\"\"\"\n", + " comps = integer_compositions(n)\n", + " print(f\"\\nCompositions of {n}: {len(comps)} total\")\n", + " print(f\"{'Composition':<20} {'Length':>6} {'Interpretation'}\")\n", + " print(\"-\" * 60)\n", + " for comp in sorted(comps, key=lambda c: (len(c), c)):\n", + " interp = []\n", + " for k in comp:\n", + " if k == 1:\n", + " interp.append(\"independent\")\n", + " elif k == 2:\n", + " interp.append(\"pairwise\")\n", + " elif k == 3:\n", + " interp.append(\"3-way\")\n", + " elif k == 4:\n", + " interp.append(\"4-way\")\n", + " elif k == 5:\n", + " interp.append(\"5-way (full)\")\n", + " else:\n", + " interp.append(f\"{k}-way\")\n", + " print(f\" {str(comp):<20} {len(comp):>4} {' → '.join(interp)}\")\n", + " return comps\n", + "\n", + "\n", + "def compare_conv4_conv5():\n", + " \"\"\"Show the scaling from conv4 to conv5.\"\"\"\n", + " print(\"=\" * 60)\n", + " print(\"COMPOSITIONAL CONVOLUTION: PARTITION ANALYSIS\")\n", + " print(\"=\" * 60)\n", + "\n", + " c4 = analyze_compositions(4)\n", + " c5 = analyze_compositions(5)\n", + "\n", + " print(f\"\\nconv4d: {len(c4)} paths (vs 1 opaque operator)\")\n", + " print(f\"conv5d: {len(c5)} paths (vs 1 opaque operator)\")\n", + " print(f\"\\nThe {len(c5)} paths for conv5d enumerate ALL ways to\")\n", + " print(f\"traverse a 5-dimensional simplex (pentachoron).\")\n", + " print(f\"Each path captures a different structural relationship.\")\n", + " print(f\"Together they form a complete basis for 5d geometry.\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " compare_conv4_conv5()\n", + "\n", + " print(f\"\\n{'='*60}\")\n", + " print(\"TESTING: DualAnchorEmbedding with conv4\")\n", + " print(f\"{'='*60}\")\n", + "\n", + " device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + " model = DualAnchorEmbedding(\n", + " embed_dim=768, hidden_dim=256, conv_n=4, bank_size=64\n", + " ).to(device)\n", + "\n", + " n_params = sum(p.numel() for p in model.parameters())\n", + " print(f\" Parameters: {n_params:,}\")\n", + " print(f\" Conv4 paths: {len(model.memory.conv_n.compositions)}\")\n", + " for i, comp in enumerate(model.memory.conv_n.compositions):\n", + " print(f\" Path {i}: {comp}\")\n", + "\n", + " # Simulate two aligned BERT embeddings\n", + " B = 16\n", + " anchor_a = F.normalize(torch.randn(B, 768, device=device), dim=-1)\n", + " anchor_b = F.normalize(torch.randn(B, 768, device=device), dim=-1)\n", + "\n", + " # Forward\n", + " output = model(anchor_a, anchor_b)\n", + " print(f\"\\n Input: anchor_a={tuple(anchor_a.shape)}, anchor_b={tuple(anchor_b.shape)}\")\n", + " print(f\" Output: {tuple(output.shape)}\")\n", + " print(f\" Output norm: {output.norm(dim=-1).mean():.4f}\")\n", + "\n", + " # Path analysis\n", + " analysis = model.memory.conv_n.path_analysis(anchor_a - anchor_b)\n", + " print(f\"\\n Path weights (learned):\")\n", + " for comp, w in zip(analysis[\"compositions\"], analysis[\"weights\"]):\n", + " print(f\" {str(comp):<20} weight={w:.4f}\")\n", + "\n", + " print(\"\\nDone.\")" + ] + }, + { + "cell_type": "markdown", + "source": [ + "# dual anchored procrustes embedding hub bert" + ], + "metadata": { + "id": "QOYafE-wP2Yr" + } + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# COMPOSITIONAL CONV EXPERIMENT: Full Pipeline\n", + "#\n", + "# 1. Extract BERT-base + ModernBERT-large pooled embeddings on CC12M captions\n", + "# 2. Procrustes-align both to shared 768-dim space\n", + "# 3. Train DualAnchorEmbedding with compositional conv4 decomposition\n", + "# 4. Losses: dual InfoNCE + pentachoron CV + running SVD alignment\n", + "# 5. Analyze: which composition paths does the model learn to weight?\n", + "#\n", + "# This tests whether integer partition decomposition of structural\n", + "# differences between two aligned models produces a useful unified\n", + "# embedding space.\n", + "# ============================================================================\n", + "\n", + "import math\n", + "import os\n", + "import time\n", + "import json\n", + "from dataclasses import dataclass\n", + "\n", + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "from tqdm import tqdm\n", + "\n", + "\n", + "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + "\n", + "print(\"=\" * 65)\n", + "print(\"COMPOSITIONAL CONV EXPERIMENT\")\n", + "print(\"=\" * 65)\n", + "print(f\" Device: {DEVICE}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STEP 1: EXTRACT EMBEDDINGS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "@dataclass\n", + "class ExtractConfig:\n", + " n_samples: int = 20000\n", + " batch_size: int = 64\n", + " max_len_bert: int = 512\n", + " max_len_modern: int = 512\n", + " min_caption_len: int = 50\n", + " cache_dir: str = \"/home/claude/comp_conv_cache\"\n", + "\n", + "ECFG = ExtractConfig()\n", + "\n", + "\n", + "def load_captions(n, min_len=50):\n", + " from datasets import load_dataset\n", + " print(f\"\\n Loading captions (n={n})...\")\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > min_len:\n", + " captions.append(cap)\n", + " if len(captions) >= n:\n", + " break\n", + " print(f\" Got {len(captions)} captions\")\n", + " return captions\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def extract_embeddings(model_name, captions, max_len, batch_size=64):\n", + " from transformers import AutoModel, AutoTokenizer\n", + " print(f\"\\n Extracting from {model_name}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " dim = model.config.hidden_size\n", + " print(f\" dim={dim}, {sum(p.numel() for p in model.parameters()):,} params\")\n", + "\n", + " all_embeds = []\n", + " for i in tqdm(range(0, len(captions), batch_size), desc=f\" {model_name.split('/')[-1]}\"):\n", + " batch = captions[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_embeds.append(pooled.cpu())\n", + "\n", + " embeds = torch.cat(all_embeds)\n", + " print(f\" Shape: {embeds.shape}\")\n", + "\n", + " del model\n", + " torch.cuda.empty_cache()\n", + " return embeds\n", + "\n", + "\n", + "def extract_or_load():\n", + " os.makedirs(ECFG.cache_dir, exist_ok=True)\n", + " bert_path = os.path.join(ECFG.cache_dir, \"bert_base.pt\")\n", + " modern_path = os.path.join(ECFG.cache_dir, \"modern_bert.pt\")\n", + " caps_path = os.path.join(ECFG.cache_dir, \"captions.json\")\n", + "\n", + " if os.path.exists(bert_path) and os.path.exists(modern_path):\n", + " print(\"\\n Loading cached embeddings...\")\n", + " bert_emb = torch.load(bert_path, weights_only=True)\n", + " modern_emb = torch.load(modern_path, weights_only=True)\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " print(f\" BERT-base: {bert_emb.shape}\")\n", + " print(f\" ModernBERT: {modern_emb.shape}\")\n", + " return bert_emb, modern_emb, captions\n", + "\n", + " captions = load_captions(ECFG.n_samples, ECFG.min_caption_len)\n", + "\n", + " bert_emb = extract_embeddings(\n", + " \"google-bert/bert-base-uncased\", captions,\n", + " ECFG.max_len_bert, ECFG.batch_size)\n", + "\n", + " modern_emb = extract_embeddings(\n", + " \"answerdotai/ModernBERT-base\", captions,\n", + " ECFG.max_len_modern, ECFG.batch_size)\n", + "\n", + " torch.save(bert_emb, bert_path)\n", + " torch.save(modern_emb, modern_path)\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " return bert_emb, modern_emb, captions\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STEP 2: PROCRUSTES ALIGNMENT\n", + "# ════════════════════��═════════════════════════════════════════════\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " \"\"\"\n", + " Align source → target space via whitened Procrustes.\n", + " Returns rotation matrix and means for inference-time alignment.\n", + " Handles dimension mismatch: projects larger down to smaller.\n", + " \"\"\"\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + "\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean\n", + " Tc = T - t_mean\n", + "\n", + " d_s, d_t = Sc.shape[1], Tc.shape[1]\n", + "\n", + " # Align dimensions\n", + " projection = None\n", + " if d_s > d_t:\n", + " _, _, Vt = torch.linalg.svd(Sc, full_matrices=False)\n", + " projection = Vt[:d_t].T\n", + " Sc = Sc @ projection\n", + " elif d_s < d_t:\n", + " pad = torch.zeros(N, d_t - d_s)\n", + " Sc = torch.cat([Sc, pad], dim=1)\n", + " projection = \"pad\"\n", + "\n", + " cos_before = F.cosine_similarity(Sc, Tc[:, :Sc.shape[1]], dim=-1).mean().item()\n", + "\n", + " # SVD for rotation\n", + " cross = Tc.T @ Sc\n", + " U, S_vals, Vt = torch.linalg.svd(cross, full_matrices=False)\n", + " R = U @ Vt\n", + "\n", + " Sc_rotated = Sc @ R.T\n", + " cos_after = F.cosine_similarity(Sc_rotated, Tc, dim=-1).mean().item()\n", + "\n", + " print(f\" Procrustes: cos {cos_before:.4f} → {cos_after:.4f}\")\n", + "\n", + " return {\n", + " \"rotation\": R,\n", + " \"source_mean\": s_mean.squeeze(0),\n", + " \"target_mean\": t_mean.squeeze(0),\n", + " \"projection\": projection,\n", + " \"cos_before\": cos_before,\n", + " \"cos_after\": cos_after,\n", + " \"d_target\": d_t,\n", + " }\n", + "\n", + "\n", + "def apply_alignment(embeddings, alignment):\n", + " \"\"\"Apply stored Procrustes alignment to embeddings.\"\"\"\n", + " x = embeddings.float() - alignment[\"source_mean\"]\n", + "\n", + " if alignment[\"projection\"] is not None:\n", + " if alignment[\"projection\"] == \"pad\":\n", + " d_t = alignment[\"d_target\"]\n", + " d_s = x.shape[1]\n", + " if d_s < d_t:\n", + " pad = torch.zeros(x.shape[0], d_t - d_s)\n", + " x = torch.cat([x, pad], dim=1)\n", + " else:\n", + " x = x @ alignment[\"projection\"]\n", + "\n", + " x = x @ alignment[\"rotation\"].T\n", + " return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STEP 3: COMPOSITIONAL CONV (from experiment file)\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def integer_compositions(n):\n", + " if n == 0: return [()]\n", + " if n == 1: return [(1,)]\n", + " result = []\n", + " for first in range(1, n + 1):\n", + " for rest in integer_compositions(n - first):\n", + " result.append((first,) + rest)\n", + " return result\n", + "\n", + "\n", + "class ConvPath(nn.Module):\n", + " def __init__(self, composition, embed_dim, hidden_dim):\n", + " super().__init__()\n", + " self.composition = composition\n", + " self.embed_dim = embed_dim\n", + " self.hidden_dim = hidden_dim\n", + "\n", + " self.steps = nn.ModuleList()\n", + " current_dim = embed_dim\n", + " for k in composition:\n", + " self.steps.append(nn.ModuleDict({\n", + " \"proj\": nn.Linear(current_dim, hidden_dim),\n", + " \"group_mix\": nn.Linear(hidden_dim, hidden_dim),\n", + " \"norm\": nn.LayerNorm(hidden_dim),\n", + " }))\n", + " self.steps[-1].k_value = k\n", + " current_dim = hidden_dim\n", + "\n", + " self.output_proj = nn.Linear(hidden_dim, embed_dim)\n", + "\n", + " def forward(self, x):\n", + " B = x.shape[0]\n", + " h = x\n", + " for step in self.steps:\n", + " k = step.k_value\n", + " h = F.gelu(step[\"proj\"](h))\n", + " if k > 1 and self.hidden_dim >= k:\n", + " n_groups = self.hidden_dim // k\n", + " if n_groups > 0 and self.hidden_dim % k == 0:\n", + " grouped = h.view(B, n_groups, k)\n", + " grouped = grouped * torch.softmax(grouped, dim=-1)\n", + " h = grouped.view(B, self.hidden_dim)\n", + " h = F.gelu(step[\"group_mix\"](h))\n", + " h = step[\"norm\"](h)\n", + " return self.output_proj(h)\n", + "\n", + "\n", + "class CompositionalConvN(nn.Module):\n", + " def __init__(self, n, embed_dim, hidden_dim):\n", + " super().__init__()\n", + " self.n = n\n", + " self.compositions = integer_compositions(n)\n", + " self.paths = nn.ModuleList([\n", + " ConvPath(comp, embed_dim, hidden_dim) for comp in self.compositions])\n", + " self.n_paths = len(self.paths)\n", + " self.path_weights = nn.Parameter(torch.ones(self.n_paths) / self.n_paths)\n", + " self.fusion = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim), nn.GELU(), nn.LayerNorm(embed_dim))\n", + "\n", + " def forward(self, x):\n", + " outputs = torch.stack([path(x) for path in self.paths], dim=1)\n", + " weights = F.softmax(self.path_weights, dim=0)\n", + " fused = (outputs * weights.view(1, -1, 1)).sum(dim=1)\n", + " return self.fusion(fused)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STEP 4: DUAL ANCHOR MODEL + TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class DualAnchorModel(nn.Module):\n", + " def __init__(self, embed_dim=768, hidden_dim=256, conv_n=4):\n", + " super().__init__()\n", + " self.embed_dim = embed_dim\n", + " self.conv = CompositionalConvN(conv_n, embed_dim, hidden_dim)\n", + " self.combine = nn.Sequential(\n", + " nn.Linear(embed_dim * 3, embed_dim),\n", + " nn.GELU(), nn.LayerNorm(embed_dim))\n", + " self.output = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim),\n", + " nn.GELU(), nn.Linear(embed_dim, embed_dim))\n", + "\n", + " def forward(self, anchor_a, anchor_b):\n", + " diff = anchor_a - anchor_b\n", + " structural = self.conv(diff)\n", + " mean = (anchor_a + anchor_b) / 2\n", + " combined = self.combine(torch.cat([structural, anchor_a, anchor_b], dim=-1))\n", + " return F.normalize(self.output(combined + mean), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRIC LOSSES\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "\n", + "def pentachoron_cv_loss(embeddings, target=0.20, n_samples=16):\n", + " B = embeddings.shape[0]\n", + " if B < 5:\n", + " return torch.tensor(0.0, device=embeddings.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=embeddings.device)[:5]\n", + " v2 = cayley_menger_vol2(embeddings[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "def svd_alignment(a, b):\n", + " A = F.normalize(a.float(), dim=-1)\n", + " B_e = F.normalize(b.float(), dim=-1)\n", + " A = A - A.mean(0, keepdim=True)\n", + " B_e = B_e - B_e.mean(0, keepdim=True)\n", + " N, D = A.shape\n", + " try:\n", + " if N < D:\n", + " S = torch.linalg.svdvals(A @ B_e.T)\n", + " else:\n", + " S = torch.linalg.svdvals(A.T @ B_e)\n", + " except Exception:\n", + " return torch.tensor(0.0, device=a.device, requires_grad=True)\n", + " return 1.0 - S.sum() / (math.sqrt(N) * D)\n", + "\n", + "\n", + "def pentachoron_cv_metric(embeddings, n_samples=200):\n", + " B = embeddings.shape[0]\n", + " if B < 5:\n", + " return 0.0\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=embeddings.device)[:5]\n", + " v2 = cayley_menger_vol2(embeddings[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0:\n", + " vols.append(v)\n", + " if len(vols) < 10:\n", + " return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "@dataclass\n", + "class TrainConfig:\n", + " epochs: int = 20\n", + " batch_size: int = 256\n", + " lr: float = 3e-4\n", + " weight_decay: float = 0.01\n", + " grad_clip: float = 1.0\n", + " # Loss weights\n", + " infonce_a_weight: float = 1.0\n", + " infonce_b_weight: float = 1.0\n", + " cv_weight: float = 0.1\n", + " svd_a_weight: float = 0.05\n", + " svd_b_weight: float = 0.05\n", + "\n", + "TCFG = TrainConfig()\n", + "\n", + "\n", + "def train():\n", + " # ── Extract ──\n", + " bert_emb, modern_emb, captions = extract_or_load()\n", + "\n", + " # ── Align both to shared 768-dim space ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PROCRUSTES ALIGNMENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " # ModernBERT-base is also 768-dim, so alignment is direct\n", + " d_bert = bert_emb.shape[1]\n", + " d_modern = modern_emb.shape[1]\n", + " d_shared = min(d_bert, d_modern)\n", + " print(f\" BERT-base: {d_bert}-dim\")\n", + " print(f\" ModernBERT-base: {d_modern}-dim\")\n", + " print(f\" Shared space: {d_shared}-dim\")\n", + "\n", + " print(f\"\\n Aligning BERT-base → shared:\")\n", + " align_bert = procrustes_align(bert_emb, modern_emb)\n", + " bert_aligned = apply_alignment(bert_emb, align_bert)\n", + "\n", + " print(f\" Aligning ModernBERT → shared:\")\n", + " align_modern = procrustes_align(modern_emb, modern_emb) # identity, but keeps API consistent\n", + " modern_aligned = apply_alignment(modern_emb, align_modern)\n", + "\n", + " # Verify alignment\n", + " N = min(5000, bert_aligned.shape[0])\n", + " cos = F.cosine_similarity(bert_aligned[:N], modern_aligned[:N], dim=-1).mean().item()\n", + " print(f\"\\n Post-alignment cosine: {cos:.4f}\")\n", + "\n", + " # Split train/val\n", + " n_val = 2000\n", + " n_train = bert_aligned.shape[0] - n_val\n", + " train_a = bert_aligned[:n_train].to(DEVICE)\n", + " train_b = modern_aligned[:n_train].to(DEVICE)\n", + " val_a = bert_aligned[n_train:].to(DEVICE)\n", + " val_b = modern_aligned[n_train:].to(DEVICE)\n", + " print(f\" Train: {n_train}, Val: {n_val}\")\n", + "\n", + " # ── Build model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"MODEL\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = DualAnchorModel(\n", + " embed_dim=d_shared, hidden_dim=256, conv_n=4).to(DEVICE)\n", + " n_params = sum(p.numel() for p in model.parameters())\n", + " print(f\" Parameters: {n_params:,}\")\n", + " print(f\" Conv4 paths: {len(model.conv.compositions)}\")\n", + " for i, comp in enumerate(model.conv.compositions):\n", + " print(f\" {i}: {comp}\")\n", + "\n", + " # ── Train ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({TCFG.epochs} epochs)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " optimizer = torch.optim.AdamW(model.parameters(), lr=TCFG.lr,\n", + " weight_decay=TCFG.weight_decay)\n", + " n_batches = n_train // TCFG.batch_size\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * TCFG.epochs, eta_min=1e-6)\n", + "\n", + " all_metrics = {\"epochs\": [], \"path_weights\": []}\n", + "\n", + " for epoch in range(TCFG.epochs):\n", + " model.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " losses = {\"total\": 0, \"nce_a\": 0, \"nce_b\": 0, \"cv\": 0, \"svd_a\": 0, \"svd_b\": 0}\n", + " metrics = {\"acc_a\": 0, \"acc_b\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, TCFG.batch_size):\n", + " idx = perm[i:i+TCFG.batch_size]\n", + " if len(idx) < 8:\n", + " continue\n", + " a = train_a[idx]\n", + " b = train_b[idx]\n", + "\n", + " emb = model(a, b)\n", + "\n", + " # 5 losses\n", + " l_nce_a, acc_a = infonce(emb, a)\n", + " l_nce_b, acc_b = infonce(emb, b)\n", + " l_cv = pentachoron_cv_loss(emb)\n", + " l_svd_a = svd_alignment(emb, a)\n", + " l_svd_b = svd_alignment(emb, b)\n", + "\n", + " loss = (TCFG.infonce_a_weight * l_nce_a +\n", + " TCFG.infonce_b_weight * l_nce_b +\n", + " TCFG.cv_weight * l_cv +\n", + " TCFG.svd_a_weight * l_svd_a +\n", + " TCFG.svd_b_weight * l_svd_b)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(model.parameters(), TCFG.grad_clip)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " losses[\"total\"] += loss.item()\n", + " losses[\"nce_a\"] += l_nce_a.item()\n", + " losses[\"nce_b\"] += l_nce_b.item()\n", + " losses[\"cv\"] += l_cv.item()\n", + " metrics[\"acc_a\"] += acc_a\n", + " metrics[\"acc_b\"] += acc_b\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Path weights\n", + " pw = F.softmax(model.conv.path_weights, dim=0).detach().cpu().tolist()\n", + "\n", + " # Val\n", + " model.eval()\n", + " with torch.no_grad():\n", + " val_emb = model(val_a, val_b)\n", + " _, v_acc_a = infonce(val_emb, val_a)\n", + " _, v_acc_b = infonce(val_emb, val_b)\n", + " v_cv = pentachoron_cv_metric(val_emb)\n", + "\n", + " summary = {\n", + " \"epoch\": epoch + 1,\n", + " \"loss\": losses[\"total\"] / d,\n", + " \"acc_a\": metrics[\"acc_a\"] / d,\n", + " \"acc_b\": metrics[\"acc_b\"] / d,\n", + " \"val_acc_a\": v_acc_a,\n", + " \"val_acc_b\": v_acc_b,\n", + " \"val_cv\": v_cv,\n", + " }\n", + " all_metrics[\"epochs\"].append(summary)\n", + " all_metrics[\"path_weights\"].append(pw)\n", + "\n", + " # Path weight string\n", + " top_paths = sorted(zip(model.conv.compositions, pw), key=lambda x: -x[1])[:3]\n", + " top_str = \" \".join(f\"{str(c)}={w:.3f}\" for c, w in top_paths)\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s \"\n", + " f\"loss={summary['loss']:.4f} \"\n", + " f\"acc_a={summary['acc_a']:.3f}/{summary['val_acc_a']:.3f} \"\n", + " f\"acc_b={summary['acc_b']:.3f}/{summary['val_acc_b']:.3f} \"\n", + " f\"cv={summary['val_cv']:.3f} \"\n", + " f\"top: {top_str}\")\n", + "\n", + " # ── Final analysis ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FINAL PATH WEIGHT ANALYSIS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " pw = F.softmax(model.conv.path_weights, dim=0).detach().cpu()\n", + " for comp, w in sorted(zip(model.conv.compositions, pw.tolist()), key=lambda x: -x[1]):\n", + " bar = \"█\" * int(w * 50)\n", + " labels = {1: \"indep\", 2: \"pair\", 3: \"3-way\", 4: \"4-way\"}\n", + " interp = \" → \".join(labels.get(k, f\"{k}-way\") for k in comp)\n", + " print(f\" {str(comp):<20} {w:.4f} {bar} ({interp})\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"TRANSITIVITY TEST\")\n", + " print(f\"{'='*65}\")\n", + " print(\" Can the learned embeddings align with models never seen during training?\")\n", + "\n", + " with torch.no_grad():\n", + " # Generate embeddings for val set\n", + " val_emb = model(val_a, val_b).cpu()\n", + "\n", + " # Direct cosine with each anchor\n", + " cos_a = F.cosine_similarity(val_emb, val_a.cpu(), dim=-1).mean().item()\n", + " cos_b = F.cosine_similarity(val_emb, val_b.cpu(), dim=-1).mean().item()\n", + "\n", + " # Midpoint check — is the embedding between the two anchors?\n", + " midpoint = F.normalize((val_a.cpu() + val_b.cpu()) / 2, dim=-1)\n", + " cos_mid = F.cosine_similarity(val_emb, midpoint, dim=-1).mean().item()\n", + "\n", + " # CV of the output space\n", + " cv_output = pentachoron_cv_metric(val_emb[:1000])\n", + " cv_anchor_a = pentachoron_cv_metric(val_a.cpu()[:1000])\n", + " cv_anchor_b = pentachoron_cv_metric(val_b.cpu()[:1000])\n", + "\n", + " print(f\" cos(output, BERT-base aligned): {cos_a:.4f}\")\n", + " print(f\" cos(output, ModernBERT aligned): {cos_b:.4f}\")\n", + " print(f\" cos(output, anchor midpoint): {cos_mid:.4f}\")\n", + " print(f\" CV output: {cv_output:.4f}\")\n", + " print(f\" CV BERT-base: {cv_anchor_a:.4f}\")\n", + " print(f\" CV ModernBERT: {cv_anchor_b:.4f}\")\n", + "\n", + " # Save\n", + " os.makedirs(ECFG.cache_dir, exist_ok=True)\n", + " with open(os.path.join(ECFG.cache_dir, \"results.json\"), \"w\") as f:\n", + " json.dump(all_metrics, f, indent=2, default=str)\n", + " torch.save(model.state_dict(), os.path.join(ECFG.cache_dir, \"model.pt\"))\n", + " print(f\"\\n Saved to {ECFG.cache_dir}/\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " train()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "af7741af1af84bd7a81fbb7d3df7a41a", + "f890ba6cefd04309be792af187a14778", + "ed379350acdf445aa85df8ffb3f7a1f1", + "ab6bb3be570143c89474b6be3082fb5b", + "cfe32c9e840c42bbbef91c80b7b8b886", + "e9a4b0f6b6eb44e5aee20c32efeea4ab", + "e932c785787047a98f376680c2928142", + "247fe9ef5fc64d25bde33a8116724852", + "34ffabfc9e164f6293bac0bbfac25b14", + "e126fdbf7d2b4b1398cf92e696170a61", + "d65e49eaa33e454486f1a7f9624d6d06", + "6b4725bd99ee4693b47742b4b95033e5", + "784f4114300447a292918ae6cc8274dc", + "d641312e61fe4238bdfd40819b512383", + "19ee6a55eace4f76aa561fe0078b3e18", + "b7152bd38a6d4388b99acd735364be70", + "4c24f92ca35b411f9a2b554523c526c0", + "91dd7aab738a4c928f93508641381d5b", + "088f7a64855744d3a195f5057e8053b2", + "72f5b45060a845eaa5776557e0ddcf47", + "8e9b2ef85269406a91b7b6e593fce474", + "507c6f4a0089432da303ef03dbbe93ea", + "b708de183007485a84fce02f0f918398", + "ce78e2bc90ad423d9180680f724d7107", + "43d9a583867d4d45a6a3291ec6bb3581", + "e4131698cac54bfba1100e8d8328aa97", + "3ae0042c5c89410295860b1ff13c7db3", + "cf806b5ba0aa40a78be556958c52d9bf", + "dad22612bf4e4958bc0e7af71d35308e", + "3121edee7f724f82bc3921482672f841", + "ce763fde05bb456cb6c9220f10757607", + "591022de240148fc88f5f02c007ce6ea", + "2f4c623aecf44199b3cddef3fcc4140e", + "216c15222c874c359070555738926819", + "3158000d5861479097563eea8c76029e", + "686b256c884a4ecdba4115c417192e14", + "876b5a1218a64365ad500bd53e1850fe", + "d15678c48b684b81b7daf2f98f161d26", + "1dba758a1c164477be0e3835c498accd", + "32286444007e41118322a8170daba070", + "177b1c0d87bb424797a8c371c7624ec8", + "70797a5f20a54b9099adf3aa23549fd1", + "a4c9cbf026004f59bb547584584ed47d", + "c28367a57ae84981acb2b4173c28cb45", + "9989d1226f6a420a9ba12cfb39eb6bd3", + "a669a3fd19fa47bfaf03ed56515c68ec", + "058e48a251d348378acbd5a51df74e86", + "ef1c325ae9d545efb78c569d1ded84c6", + "6025ac339fd34cd38fc444cdf44299f3", + "49475ec1c57e4da3a4c149de7a68b787", + "a8001a46a8e34afca224c48e13a751e8", + "7ed9424968cd48c69762e2c2796293fa", + "964f870da0dd41d9879cda99e2583c64", + "c2e096c6cc174454a9c7ea5b0c611ed9", + "ccac7943ff9b42eba4fefc79cf8e93aa", + "249e4535faaa4866a374f6fea46b57a2", + "0ef63ff1dc114ebeb4e5f3388e396faa", + "57d2d9dcbf7c497b9b07312688914a75", + "0c8bb082dd924f43809a544fbf6e22ef", + "b77cb0b275d8476bae59fc6c17d47115", + "dc088f102c904774ac170cd157a18cf0", + "03599554e1214fa59c53e3a65075ff55", + "0e84ed64f2f44ed69213374cc614d494", + "cbc3ef4fc6e1461db49506d61eee4f21", + "55e78b09dde14e1182e27ff963b3f472", + "767d6e11ebc743b68d23e62960686e61", + "5a91836f467e46959f1b80db2ca6afea", + "b840c5497d154057b7eebe02f47ccc51", + "1672c7add2d544428851a21ed996dd0a", + "4dc0c89c1a724f159f1ca1473dac9f5f", + "098c25ce400b4293980b2bdbcc9f2bf9", + "23f8dc075ee94518927e405921fb5d4b", + "098c1f83a8f24c41ab4cd7e229a56754", + "6a3947c2b7c744bfbed33e605327e1e7", + "c984888d8311469ebf2c66d83c98c749", + "c665765cb32942448634fb62540aff20", + "913f6311e9bf4058a2d301295e01aae7", + "eed503f995d74c449acfb217ce0acd7d", + "584923fc781040288e04c2df23ba6207", + "21209e6cc4af496c96a918ee64f7e114", + "7c82222cb6e54855935571450a66d72c", + "2b9bf9f8d67f4f31ba5f592f68431da8", + "06a18d9b194f4d5783ba45f2b674cded", + "7f237b344f734beb9dbe005c33a553dd", + "52cdc94fe2b84df88e54a7c34f5fd38f", + "dfa6dce60ea44b3d8b3a264b90684fc0", + "261b5f47bc1748b98cca6125a96fc8e0", + "c8efdf5df992463e9cca11c28186d33d", + "2c869b1b80574876bb99de5b3535fd2c", + "ba8969ad55e44c638e115599042a7b37", + "8a4fbfc103be43c28d430a40de7d2213", + "e1ae0da775fb4af1bbcb3387c6c3deac", + "e945c3c437ca4ac69901538823b69203", + "698bca7a9f6c4b0c81e047e49a9aa046", + "45a1b40fa4a0460abc61e6d82ce14a4c", + "5c6ac9d91f4245c6b96ad0a5d9102e41", + "bb9870743c2c49699d04fc1802b72577", + "78b2e7158eb94aedba001fbd14fe98c3", + "b220cdda21a94f2dbaa3b627fbeb78ee", + "af2a792cfc4a4886afb14f23a69e8e32", + "017d834170154d6caa6fe3fbcad87cc8", + "f94b989f3400401fb93db36bb4ae1e43", + "ef1197d0a7d24adcafb7f62814f2fa64", + "4e22cb7b7eb8437d94a5f3cd84c00cad", + "82fdb2e5fc3540c2adcd4717520b4f05", + "2e55bc310fd94b5b88a11ed8844cf989", + "e57da4bd40da4427a224607b5a17fcdf", + "8c40cdfb5ae544dea8f1458da2d35cb1", + "158bc2ca4b304e6e94fcd839a30d47f4", + "f5e3d0180b7f418693ca40d042674d97", + "a085e226619c41f6853fef4568929453", + "4fc62e7e1e19420aa7974190a48b76ac", + "f419ed984fe14816bf27f313ae93b7c7", + "2cfb05e56c964bbfae9b4eab1935c4e7", + "ca7dc6d7370347699c4fdf7a210aa898", + "51cc7cb998b24fa5a62722a363f18fd2", + "62f7292135d24aeeb9eb4ca2f8fb93e9", + "cc3251b8ba30447abf38e3d1804e4c45", + "0e28de4e717e4d26a003b1da684a9b50", + "15878b16affc454d8edf48f9ade837bf", + "d1952bd013b742b988caf768547b648f", + "d8491c9e16c44cb686845d39f8ac392d", + "772fbd960f124a39b87531b2661ba5b8", + "4b73d09efe854ed99bf962c1f6791d4b", + "425902f67d044118a72cefcb35a4a7da", + "39804dbc2f224ff49fe4dbaba666bdab", + "cce9161655c943538a5c485a6c5aeff4", + "044ce8c81d2a4d67beae9f945068e5cb", + "4ca65781df6242a6ab2b1f72a1de9f4a", + "8158cfa78ddb4cf6a6693cb026347a6a", + "fbdf4d1087614c03af3037b03208df82", + "c4dbefe5d8e6410db7b2228d897766f6", + "91a81ac85cd04855b64b721830cb1ac2", + "6ceef1cd00314bdaa54537cb4ccd8634", + "6a345a7d08b54545a02a4bd706705d9a", + "02e8c91e46af4e6bbab16b9a80b675a8", + "632c821a403d4cbdaa18e9c96848bdfc", + "108db55efdf049efb551e9397c3d0072", + "c7d4f9bea5474958ad5975451c65fdd4", + "8db8d51ee82f408793ff80a62c7654eb", + "20f3700205444e24ae3562944a5dadad", + "dec81f3e632045ef9a2b91ec1febb795", + "a23face7e21b44918c9f1edea5053c08" + ] + }, + "id": "fs6DkaDpPmbK", + "outputId": "7eb718f5-55b7-48ae-d173-ed93eafdf6b8" + }, + "execution_count": 2, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "COMPOSITIONAL CONV EXPERIMENT\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + " Loading captions (n=20000)...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "README.md: 0.00B [00:00, ?B/s]" + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "af7741af1af84bd7a81fbb7d3df7a41a" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " Got 20000 captions\n", + "\n", + " Extracting from google-bert/bert-base-uncased...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "config.json: 0%| | 0.00/570 [00:00 min_len:\n", + " captions.append(cap)\n", + " if len(captions) >= n:\n", + " break\n", + " print(f\" Got {len(captions)} captions\")\n", + " return captions\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def extract_one(model_name, short_name, captions, max_len, batch_size):\n", + " from transformers import AutoModel, AutoTokenizer\n", + " print(f\"\\n Extracting: {short_name} ({model_name})...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " dim = model.config.hidden_size\n", + " print(f\" dim={dim}, {sum(p.numel() for p in model.parameters()):,} params\")\n", + "\n", + " all_emb = []\n", + " for i in tqdm(range(0, len(captions), batch_size), desc=f\" {short_name}\"):\n", + " batch = captions[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " # Mean pool\n", + " hs = out.last_hidden_state\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (hs * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + "\n", + " emb = torch.cat(all_emb)\n", + " print(f\" Shape: {emb.shape}\")\n", + " del model\n", + " torch.cuda.empty_cache()\n", + " return emb\n", + "\n", + "\n", + "def extract_all():\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " caps_path = os.path.join(CFG.cache_dir, \"captions.json\")\n", + "\n", + " # Check cache\n", + " all_cached = all(\n", + " os.path.exists(os.path.join(CFG.cache_dir, f\"{short}.pt\"))\n", + " for _, short in MODELS)\n", + "\n", + " if all_cached:\n", + " print(\"\\n Loading cached embeddings...\")\n", + " embeds = {}\n", + " for _, short in MODELS:\n", + " embeds[short] = torch.load(\n", + " os.path.join(CFG.cache_dir, f\"{short}.pt\"), weights_only=True)\n", + " print(f\" {short}: {embeds[short].shape}\")\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " return embeds, captions\n", + "\n", + " captions = load_captions(CFG.n_samples, CFG.min_caption_len)\n", + "\n", + " embeds = {}\n", + " for model_name, short in MODELS:\n", + " emb = extract_one(model_name, short, captions,\n", + " CFG.max_len, CFG.batch_size)\n", + " # Handle dimension mismatch — ALBERT-base-v2 is 768 but some\n", + " # models might differ. Pad/truncate to 768.\n", + " if emb.shape[1] != 768:\n", + " print(f\" Adjusting {short} from {emb.shape[1]} to 768\")\n", + " if emb.shape[1] < 768:\n", + " emb = F.pad(emb, (0, 768 - emb.shape[1]))\n", + " else:\n", + " emb = emb[:, :768]\n", + " embeds[short] = emb\n", + " torch.save(emb, os.path.join(CFG.cache_dir, f\"{short}.pt\"))\n", + "\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " return embeds, captions\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# PROCRUSTES ALIGNMENT (all to first model's space)\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean\n", + " Tc = T - t_mean\n", + "\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + "\n", + " N_s, D = Sc.shape\n", + " try:\n", + " if N_s < D:\n", + " cross = Tc.T @ Sc\n", + " else:\n", + " cross = Tc.T @ Sc\n", + " U, _, Vt = torch.linalg.svd(cross, full_matrices=False)\n", + " R = U @ Vt\n", + " except Exception:\n", + " R = torch.eye(D)\n", + "\n", + " Sc_rot = Sc @ R.T\n", + " cos_after = F.cosine_similarity(Sc_rot, Tc, dim=-1).mean().item()\n", + "\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"target_mean\": t_mean.squeeze(0),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "\n", + "def apply_align(emb, alignment):\n", + " x = emb.float() - alignment[\"source_mean\"]\n", + " return x @ alignment[\"rotation\"].T\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# COMPOSITIONAL CONV5D\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def integer_compositions(n):\n", + " if n == 0: return [()]\n", + " if n == 1: return [(1,)]\n", + " result = []\n", + " for first in range(1, n + 1):\n", + " for rest in integer_compositions(n - first):\n", + " result.append((first,) + rest)\n", + " return result\n", + "\n", + "\n", + "class ConvPath(nn.Module):\n", + " def __init__(self, composition, embed_dim, hidden_dim):\n", + " super().__init__()\n", + " self.composition = composition\n", + " self.hidden_dim = hidden_dim\n", + " self.steps = nn.ModuleList()\n", + " current_dim = embed_dim\n", + " for k in composition:\n", + " self.steps.append(nn.ModuleDict({\n", + " \"proj\": nn.Linear(current_dim, hidden_dim),\n", + " \"mix\": nn.Linear(hidden_dim, hidden_dim),\n", + " \"norm\": nn.LayerNorm(hidden_dim),\n", + " }))\n", + " self.steps[-1].k_value = k\n", + " current_dim = hidden_dim\n", + " self.out = nn.Linear(hidden_dim, embed_dim)\n", + "\n", + " def forward(self, x):\n", + " B = x.shape[0]\n", + " h = x\n", + " for step in self.steps:\n", + " k = step.k_value\n", + " h = F.gelu(step[\"proj\"](h))\n", + " if k > 1 and self.hidden_dim >= k and self.hidden_dim % k == 0:\n", + " g = h.view(B, self.hidden_dim // k, k)\n", + " g = g * torch.softmax(g, dim=-1)\n", + " h = g.view(B, self.hidden_dim)\n", + " h = F.gelu(step[\"mix\"](h))\n", + " h = step[\"norm\"](h)\n", + " return self.out(h)\n", + "\n", + "\n", + "class CompConv5d(nn.Module):\n", + " def __init__(self, embed_dim, hidden_dim):\n", + " super().__init__()\n", + " self.compositions = integer_compositions(5) # 16 paths\n", + " self.paths = nn.ModuleList([\n", + " ConvPath(c, embed_dim, hidden_dim) for c in self.compositions])\n", + " self.weights = nn.Parameter(torch.ones(16) / 16)\n", + " self.fusion = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim), nn.GELU(), nn.LayerNorm(embed_dim))\n", + "\n", + " def forward(self, x):\n", + " outs = torch.stack([p(x) for p in self.paths], dim=1)\n", + " w = F.softmax(self.weights, dim=0)\n", + " return self.fusion((outs * w.view(1, -1, 1)).sum(1))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# FIVE-ANCHOR MODEL\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class FiveAnchorModel(nn.Module):\n", + " def __init__(self, embed_dim=768, hidden_dim=256, n_anchors=5):\n", + " super().__init__()\n", + " self.n_anchors = n_anchors\n", + " self.conv5 = CompConv5d(embed_dim, hidden_dim)\n", + "\n", + " # Combine all 5 anchors + structural signal\n", + " self.combine = nn.Sequential(\n", + " nn.Linear(embed_dim * (n_anchors + 1), embed_dim),\n", + " nn.GELU(), nn.LayerNorm(embed_dim))\n", + " self.output = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim),\n", + " nn.GELU(), nn.Linear(embed_dim, embed_dim))\n", + "\n", + " def forward(self, anchors):\n", + " \"\"\"\n", + " anchors: list of 5 tensors, each (B, 768)\n", + " Returns: (B, 768) unified embedding\n", + " \"\"\"\n", + " # Structural signal: sum of all pairwise differences\n", + " # This captures the full 5-way structural relationship\n", + " structural = torch.zeros_like(anchors[0])\n", + " for i in range(self.n_anchors):\n", + " for j in range(i+1, self.n_anchors):\n", + " structural = structural + (anchors[i] - anchors[j])\n", + "\n", + " # Decompose through conv5d\n", + " decomposed = self.conv5(structural)\n", + "\n", + " # Mean of all anchors\n", + " mean_anchor = sum(anchors) / self.n_anchors\n", + "\n", + " # Combine: decomposed + all anchors\n", + " combined = self.combine(torch.cat([decomposed] + anchors, dim=-1))\n", + "\n", + " return F.normalize(self.output(combined + mean_anchor), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# LOSSES\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "\n", + "def cv_loss(emb, target=0.20, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "def svd_align(a, b):\n", + " A = F.normalize(a.float(), dim=-1)\n", + " B_e = F.normalize(b.float(), dim=-1)\n", + " A = A - A.mean(0, keepdim=True)\n", + " B_e = B_e - B_e.mean(0, keepdim=True)\n", + " N, D = A.shape\n", + " try:\n", + " S = torch.linalg.svdvals(A @ B_e.T) if N < D else torch.linalg.svdvals(A.T @ B_e)\n", + " except Exception:\n", + " return torch.tensor(0.0, device=a.device, requires_grad=True)\n", + " return 1.0 - S.sum() / (math.sqrt(N) * D)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def train():\n", + " # ── Extract ──\n", + " embeds, captions = extract_all()\n", + "\n", + " # ── Align all to first model's space ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PROCRUSTES ALIGNMENT (all → bert space)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref_name = MODELS[0][1] # bert\n", + " aligned = {}\n", + " align_info = {}\n", + "\n", + " for _, short in MODELS:\n", + " if short == ref_name:\n", + " aligned[short] = embeds[short].float()\n", + " align_info[short] = {\"cos_before\": 1.0, \"cos_after\": 1.0}\n", + " print(f\" {short:10s}: reference (identity)\")\n", + " continue\n", + "\n", + " info = procrustes_align(embeds[short], embeds[ref_name])\n", + " aligned[short] = apply_align(embeds[short], info)\n", + " align_info[short] = info\n", + " print(f\" {short:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}\")\n", + "\n", + " # Pairwise cosines after alignment\n", + " print(f\"\\n Pairwise cosines (post-alignment):\")\n", + " names = [s for _, s in MODELS]\n", + " for i in range(len(names)):\n", + " for j in range(i+1, len(names)):\n", + " cos = F.cosine_similarity(\n", + " aligned[names[i]][:5000], aligned[names[j]][:5000], dim=-1\n", + " ).mean().item()\n", + " print(f\" {names[i]:8s} ↔ {names[j]:8s}: {cos:.4f}\")\n", + "\n", + " # Split\n", + " n_val = 2000\n", + " n_train = CFG.n_samples - n_val\n", + " train_data = {k: v[:n_train].to(DEVICE) for k, v in aligned.items()}\n", + " val_data = {k: v[n_train:].to(DEVICE) for k, v in aligned.items()}\n", + " anchor_names = [s for _, s in MODELS]\n", + " print(f\"\\n Train: {n_train}, Val: {n_val}\")\n", + "\n", + " # ── Model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"MODEL: FiveAnchorModel + CompConv5d\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = FiveAnchorModel(768, CFG.hidden_dim, 5).to(DEVICE)\n", + " n_params = sum(p.numel() for p in model.parameters())\n", + " print(f\" Parameters: {n_params:,}\")\n", + " print(f\" Conv5d paths: {len(model.conv5.compositions)} (2^4 = 16)\")\n", + "\n", + " # ── Train ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({CFG.epochs} epochs)\")\n", + " print(f\" 5 InfoNCE + 5 SVD + 1 CV = 11 losses\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " optimizer = torch.optim.AdamW(model.parameters(), lr=CFG.lr,\n", + " weight_decay=CFG.weight_decay)\n", + " n_batches = n_train // CFG.train_batch\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * CFG.epochs, eta_min=1e-6)\n", + "\n", + " all_metrics = {\"epochs\": [], \"path_weights\": [], \"alignment\": align_info}\n", + "\n", + " for epoch in range(CFG.epochs):\n", + " model.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " accs = {n: 0 for n in anchor_names}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, CFG.train_batch):\n", + " idx = perm[i:i+CFG.train_batch]\n", + " if len(idx) < 8: continue\n", + "\n", + " batch_anchors = [train_data[name][idx] for name in anchor_names]\n", + " emb = model(batch_anchors)\n", + "\n", + " # 5 InfoNCE losses + 5 SVD losses\n", + " loss = torch.tensor(0.0, device=DEVICE)\n", + " for k, name in enumerate(anchor_names):\n", + " l_nce, acc = infonce(emb, batch_anchors[k])\n", + " l_svd = svd_align(emb, batch_anchors[k])\n", + " loss = loss + CFG.nce_weight * l_nce + CFG.svd_weight * l_svd\n", + " accs[name] += acc\n", + "\n", + " # CV loss\n", + " l_cv = cv_loss(emb)\n", + " loss = loss + CFG.cv_weight * l_cv\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(model.parameters(), CFG.grad_clip)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " total_loss += loss.item()\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Val\n", + " model.eval()\n", + " with torch.no_grad():\n", + " val_anchors = [val_data[name] for name in anchor_names]\n", + " val_emb = model(val_anchors)\n", + " val_accs = {}\n", + " for k, name in enumerate(anchor_names):\n", + " _, va = infonce(val_emb, val_anchors[k])\n", + " val_accs[name] = va\n", + " val_cv = cv_metric(val_emb)\n", + "\n", + " # Path weights\n", + " pw = F.softmax(model.conv5.weights, dim=0).detach().cpu().tolist()\n", + "\n", + " # Top 3 paths\n", + " sorted_paths = sorted(zip(model.conv5.compositions, pw), key=lambda x: -x[1])\n", + " top3 = \" \".join(f\"{str(c)}={w:.4f}\" for c, w in sorted_paths[:3])\n", + "\n", + " # Acc summary\n", + " acc_str = \"/\".join(f\"{accs[name]/d:.3f}\" for name in anchor_names)\n", + " vacc_str = \"/\".join(f\"{val_accs[name]:.3f}\" for name in anchor_names)\n", + "\n", + " summary = {\n", + " \"epoch\": epoch + 1, \"loss\": total_loss / d,\n", + " \"train_accs\": {n: accs[n]/d for n in anchor_names},\n", + " \"val_accs\": val_accs, \"val_cv\": val_cv,\n", + " }\n", + " all_metrics[\"epochs\"].append(summary)\n", + " all_metrics[\"path_weights\"].append(pw)\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} \"\n", + " f\"acc={acc_str} val={vacc_str} cv={val_cv:.3f} top: {top3}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # FINAL ANALYSIS\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FINAL PATH WEIGHT ANALYSIS (16 paths)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " pw = F.softmax(model.conv5.weights, dim=0).detach().cpu()\n", + " labels = {1: \"indep\", 2: \"pair\", 3: \"3-way\", 4: \"4-way\", 5: \"5-way\"}\n", + " for comp, w in sorted(zip(model.conv5.compositions, pw.tolist()), key=lambda x: -x[1]):\n", + " bar = \"█\" * int(w * 80)\n", + " interp = \" → \".join(labels.get(k, f\"{k}\") for k in comp)\n", + " print(f\" {str(comp):<25} {w:.4f} {bar} ({interp})\")\n", + "\n", + " # Weight spread\n", + " pw_arr = pw.numpy()\n", + " print(f\"\\n Weight spread: min={pw_arr.min():.4f} max={pw_arr.max():.4f} \"\n", + " f\"std={pw_arr.std():.6f} range={pw_arr.max()-pw_arr.min():.6f}\")\n", + " print(f\" Uniform would be: {1/16:.4f} = 0.0625\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"CONSENSUS GEOMETRY\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model.eval()\n", + " with torch.no_grad():\n", + " val_anchors = [val_data[name] for name in anchor_names]\n", + " val_emb = model(val_anchors).cpu()\n", + "\n", + " # Cosine to each anchor\n", + " for name in anchor_names:\n", + " cos = F.cosine_similarity(\n", + " val_emb, val_data[name].cpu(), dim=-1).mean().item()\n", + " print(f\" cos(output, {name:10s}): {cos:.4f}\")\n", + "\n", + " # Cosine to geometric centroid\n", + " centroid = F.normalize(sum(val_data[n].cpu() for n in anchor_names) / 5, dim=-1)\n", + " cos_cent = F.cosine_similarity(val_emb, centroid, dim=-1).mean().item()\n", + " print(f\" cos(output, centroid): {cos_cent:.4f}\")\n", + "\n", + " # CV comparison\n", + " out_cv = cv_metric(val_emb[:1000])\n", + " print(f\"\\n CV output: {out_cv:.4f}\")\n", + " for name in anchor_names:\n", + " acv = cv_metric(val_data[name].cpu()[:1000])\n", + " print(f\" CV {name:10s}: {acv:.4f}\")\n", + "\n", + " # Pairwise cosines between output and each anchor\n", + " print(f\"\\n Equidistance check (should be ~equal):\")\n", + " cosines = []\n", + " for name in anchor_names:\n", + " cos = F.cosine_similarity(\n", + " val_emb, val_data[name].cpu(), dim=-1).mean().item()\n", + " cosines.append(cos)\n", + " print(f\" Range: {max(cosines)-min(cosines):.6f}\")\n", + " print(f\" Std: {np.std(cosines):.6f}\")\n", + "\n", + " # Save\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " with open(os.path.join(CFG.cache_dir, \"five_results.json\"), \"w\") as f:\n", + " json.dump(all_metrics, f, indent=2, default=str)\n", + " torch.save(model.state_dict(), os.path.join(CFG.cache_dir, \"five_model.pt\"))\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " train()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 0, + "referenced_widgets": [ + "e2a70dcb8e624d7ca612ba7bb8c32807", + "ea3eb28389e64a86af508ad6f2d1a6ac", + "23f1cccb23564eba9e1e53d82ccf3c78", + "36002b606f774b2eb2a9e158acf0f197", + "d32f2d541d34419e9e2464c2ac664341", + "061f46eb38264ecebdad782966a10c47", + "bebfb41784e1440cb605c43325394e63", + "e97b733720ea47a3a68ec22e43f9b3e6", + "56454f346b4047c994e068d2de5347b1", + "52d8c3faf2484db2881201ae857be33b", + "2edd21ec2d664f039ba2beb7f91d90d5", + "83d590542cd046c1973e3134f758a50f", + "3a7256139434453eb33e1feaef040346", + "aef3afc5e6ab47969a396fab50f3f9cb", + "a1e2a7e6f3b24ec5af32a65096f7e77c", + "8c70676303854eb78eb23f3327774a0f", + "231321a4c42c4cf1b6eacc1b3fdcd45b", + "f1183aa32f974ddd97dcb1943b2575c7", + "3b6cad5748534d88a198a58224d352a1", + "fd3edc5d5df542639132bdb5bf15dd37", + "181cadcf3bb6486c9710946daf92f28a", + "0f4c4ff3caed46c1a1ba1dc3fda5357c", + "62b084d4045f495f956b8c77b53d979b", + "b95b061016fc466c92c039fde8d782ad", + "0dc3db85b0fb444da5175f161fb4cb64", + "d01aea89ae924779b6e649e7b75c2188", + "0f73b0865aed4866a5b138fa2891033f", + "634a0196709a4268b2b2dfc035ef2633", + "2e7bb615b7ad44b286ea675a17c6de1d", + "5472f2e989744382ae0347b545ee25d6", + "fbba8d61e96f49c9aec2cde5ab51a8f9", + "86940e9440ec464eb5d92496b6a214e5", + "7f22f17245134bf1b1c2518becce877e", + "f2948dca39ba4ecc9d98b1102133ed34", + "c100049742444cf9b85e175b6d2d7430", + "9b3127ed91eb4cfc8785e580c50f4c23", + "4660729f84194c52a80cbf6f29892d56", + "4c20a430b6da43a1be4080bb8461490e", + "f5200448b77f4964a9956b75988ab47a", + "61b0266940074f74bcad883c522af1c0", + "4633d4dcc2b945eeb75c14a418d4954f", + "41d0cc3eab034873b41711d6464143b4", + "ed42298f6a33403bbc9f6877908ef4b8", + "26cb536be18c49669d547ef543abd23d", + "cb5fb6c625ba49e9ab6a03a709725662", + "f9385bf5b575401985f0d05ba3348ef7", + "9392f75ede4a44f7bf880e2e0210783e", + "d594a78f634e4d76940590350ec15e43", + "70e97d70ab3d4acba3326bb2e47151b4", + "5cf478922c4447f5acd97aaf233c0727", + "ff3c65ba985b4e1f8acf3f95061c54f6", + "28ed8de161b2450798b3a1fa04d7e82b", + "0cd1f7c0b786403d8ba40a8008e50517", + "d94e7581fcab4474bd751eb3a8897f3c", + "2df3c52aec3440c08e56dd9ccbdf2138", + "80433f6dfd8d49949a7070b6cf65edb1", + "9101e148e3db460d89e4d644f612fb48", + "79261c5003774f7e8aa78536da2396e6", + "b6455e90ba3e4723aefc133e2fd4fd1e", + "17dcf7cbb6904d44a98eef1a232c046d", + "8f683fa10a7643d0bc2acd3bcd11c6e4", + "e37af08ec462491e8460cec81f75f645", + "31e200e2295a432a9e2f34df9b1c5d8f", + "d8c1777a148740dc9178bd2ce79484b3", + "db4cafd36e9340bead8b71d66558731c", + "a6f32e2206664bfc9bc10b45003bfc7c", + "f3a9156ccae24f7982bc2331d9f4efbd", + "889099da7a914f989d176034fc542749", + "618ce6d77e1c49f59294492982f9b8f7", + "4fc77c8e89a44eb59f69dd4507f83500", + "258e203e3ed3448f89edbb13876d586c", + "de0171a8b863438d866eb33cb10bd5a2", + "0453764ff6594970859447f2b68cfdef", + "64e72ab3ec174c2f905f65858ceba074", + "bc59bc91a14c4172aca65ef536d9cb05", + "611fc78d12e2420e932c5a45d96744a2", + "cd79f5e46fd94a45a7d7f4c0aac8959b", + "3c245882f54848338fe6b071b536b713", + "8660f44e8efd473e8c1c695d5b860d49", + "557a696649b04a3295766bc64c092692", + "187b10e26325482c96824ba7417d3a6a", + "a69ae42a7ecb4db8b409acbe40ba7c0a", + "cae4a557f8ca444a98fd91a956212b5f", + "d4c14d6d688a45e1b0c9a45f5eea0006", + "55c9a9d615fe43cbae080a3aa74522cb", + "3b8261809aaf422caa65a8a1aba6f726", + "d1cea2635e02449ba64d7533c7c0e772", + "76dbe1b8b6734ee9b663b64a9756a85f", + "d21ad66f0bee4ed7a715cf5be3c3de13", + "28deaa25dee94e8cb665323cadebb3ab", + "c9bf3dc381214fc680a67677d38120ca", + "769e0cec00484eddba520ee2573c20fb", + "c99b0229a251429a867f67786c345126", + "b1ba274cdf454d3e84c4f992b3804cfb", + "26bae17f338e4da3a11e7e7a256541fa", + "f098d81430dc49eda2dd364cb0ad4166", + "59cee6f235a14eecbb01578cfa73e374", + "b23fe17ac64049c78fd56fe31da7f47b", + "5ea02354c8d844bfa0e64383b4a48003", + "bf0fc650cd764e4bb91845df43652588", + "a127908ce3f04fe2aeef8e3c34eeff79", + "178af9e08e184caa8a6eba114b5412d8", + "ed84b2dfdb7a410ba25b50e639a78e23", + "3ffbfc7e2d69438cb7ba83ffed391be0", + "dad41b0099f44d50875daa47b2d7db6f", + "bbae2ea3bfc94322a3fc960e9ff58bc2", + "70a74bf0df9b45b0990db8cbbcf9d0af", + "2ce31be0bebc4b44b8404d4e81ce9c3c", + "043827c1bed044d9a4fe14a353cccbfc", + "a97d5f96f2af4b42ae08503db12c4ea4", + "a7326715179e43a78b1efe16165988d1", + "edb27c6faf0c4636a6e7e713d700cec3", + "ef69b38c0e1b4242b2a7d2d3e9382eaf", + "9b96cb040afa4bdc9b5d85c1a20cd592", + "86defdd6dd354c068eda620264f2b2bb", + "73ee733a1afb4674bd46ba582d504312", + "c45f9a287be24e8aaa41a93553621038", + "d4beb1b020a442ecb7bbe5f144eb9ec1", + "722f0c93b55a42ebbce9b05515155b9a", + "f41ab9f3f1a843a0a9d4922f3c891755", + "97b07779f8634852b3dd2a9ca33aefce", + "29b4b025d9e749cb8828a088a819a78f", + "67192e1b2c8f4996aee8a7ac87730bfa", + "1a7b34cddb3347b8b37bdc2603646055", + "331a16de953840dca34c3370a2907830", + "748f5114da474866b1b737b0010f0978", + "9e2659f5996949918dd84b9f7ed010ac", + "9e306c886520406ab752fedfa23e7758", + "482069bc1b1d4fe684f05c73a208a240", + "d055c436b5124b4295a9ae3ff1129462", + "7d14322b6bb64d8aacd68cfdfd543149", + "b522e34d74d1459cac9fe8218a1f3287", + "e6bdea9e40b44959a5c58a9e279af2d0", + "b38a2df501c4486da4b0f1ba9b1819ec", + "0cccd0189a4a4268b297b76448ba7bdb", + "10d4198a08ef4995877e63bb3fa528a8", + "2b78c4d3b6c74687a960d7dad357a132", + "561bd0ea372d4f2987c340894fda2f6f", + "e8dbe71c495d4d2ea289117544889a66", + "4395d639c16d4c6caaa274f2161bef23", + "dea61b9776f34c099930b4d0b5dd4032", + "9da65bd197794b81a12b6ab59a36577b", + "c22e76554ca149288e3bd7c731dcc6af", + "b34f02689eb74f10b587d15bd355c4f0", + "ea1d6d9fc1f14dfd9bd766ced0b7408f", + "7b6e7f8037544f9a8bd6987b20e26ce6", + "cc0d6b6e28414bba9455ae019e1cf077", + "9902f2dc7f8043ca8a54a751a11e16a5", + "248e37ccd3b04d58be04cfd1f743f00f", + "d30dc98d3c1543f1ab4576d42c884820", + "d58ef9fc339a4c078f59a48dccf14269", + "0963bfd1678342bfad73f57615dc3253", + "a183e56ca7704eecaa2ee15ca5a1c62b", + "995fb042fcc74a2aae45c52fda9b1f8b", + "e396f3c394284383a8c02fb5509dab40", + "78d7e796eb934b95bf408f2de7ee6183", + "b3edbb4a0ffd42f1a7422fa9b0c598e8", + "1795f93e690748b99caaa90d805b72d2", + "e97177add3844f72a71a624c71403309", + "bced4455899849e9b549ee5242389e0d", + "1915dfeb013a4cbebd14f0c244bf0d59", + "95cab53c46a240fcbe9966bda5ce26ce", + "0f99cb8305ac4246b8b74588f726ac33", + "2ce464af80ab4c25a4d15a74569c12ad", + "f3826f5dbd2846baa00931b8c559b5dc", + "7fe531947a98458fb61ad5ccece362ad", + "ff0ce2d8a69b4024aee9313567801eac", + "520c5393a85c40f5a0905c3d9fad6c9c", + "e872a1f8826041c7b91e152bd2b70831", + "5e2a2b6f85bc4688bd5524ce74618239", + "ec6422eeba5e43a68d97ca9a480b04f1", + "85b7d5f09eab4f8898d39d1125908c97", + "6b594c0e28f7472ba5d8449eb2a814b3", + "9225d3431b9e4bf8878670033cd37048", + "d50f3aa199ac448e9e5af9f6de771990", + "38164c5851694c978f4cae3599b46b89", + "a66cbab44e1f44dc96447a4f7b00f798", + "abc873dbcbe64a81bb0fee6608294bdf", + "8f942478ee1e457cb61e07abba8edc70", + "bcb36fb962584748836b7ca776775d2a", + "f07ab6e1039c49d18d66a217afa29897", + "3241865cbc6c442a95d4d38e82aa25ac", + "4893c83a88044ca39c5f00b686c09a03", + "d34cce7e6ca44bd0a3664df642a1f7dc", + "ece74c977a73489f9efdec2fa7e778ba", + "8047d5dbbcad42aa8969cd6aa2eca828", + "5c2dfffdc8b04d1885d9ce1f02a7d0d1", + "c3e0a5e0dc2547fd96e384b4142094fb", + "9e0098e4f054480fae8dccb70eb5fc16", + "dda522f33a23466db417679e65673e24", + "aeafa57dec3442ccb07287952e31bf16", + "bd51f8731dce43a0a73da9133037a9f7", + "9c98f555b91d4edd9e1e84043e60edc7", + "0e381b6c796443a0a5734254bbe09ea5", + "95c3a67d20a74da58e619915e17d25ba", + "ee5c30553b3e48588295b90295d7943a", + "23f17bc680d34ad194a26b8b93693e1b", + "d47da607c91640ed93859e5be6137d01", + "77727da895e64e33afb03b21abf632a9", + "f9d1285d6c9f47d4a5586a4002b3461f", + "8b8b9f5e4dc741da8fcdaa667fe4cffa", + "2e7d00e6737d419daa31f38427df40d2", + "40dc7545a0b547a396d666457057cc34", + "760719f9e92349ca9360e900e589271e", + "925232a874d840fb8ae504c5cdcbec2f", + "6bce40c76023457281f6838b3ba74f69", + "54538f8399184581837b414057445b49", + "2ab7eee807c746ffa44fdff3fec9cd2f", + "a81bad4243984c99a1f2fbe0355f72c3", + "f6585ffef775480c867f8ac095bdc292", + "c70e8217531d4778a4d7341a8604e514", + "afe1af71c4dd44929c2a1b2a8d02f6e9", + "aede4d97d7ef4e899ec63d29b8fbfe9c", + "2a4db79fbbaa4e7b885c0c8dfbf3a9f6", + "00f59dabbd1a4377ba0b0bbebd7523a6", + "49342ca5408f4ba1985b64e307802b99", + "68988518c92545c0a2fa1900e6ad30e1", + "2ccded8d8b0a4317947c3f04813bdcc0", + "345dc059805148d5ae130956fe755da8", + "b9767ace4e39498188862d285e1c7234", + "346ce9b77d454356b569ec3383f1492f", + "4c3d20d7ede34cb0934bc1ca0b7ec4bd", + "18449ba42d6f4997aecdcf7ccc0253c7", + "e3c98c90f79f4917afbcb712e2408839", + "831f5133f327465da02e6ac03a77e03e", + "9baf758c7b324cba91f3af4d9107d944", + "b639b1a63b59473daad7deced5c3d60c", + "933a7bd9df2c43a3b71ecd02d897aa9f", + "23593ce9f71d4424b7e089d78111baab", + "25b604ffaf9d49878f4326ffcc90fc11", + "0edd02992610457397183dcfe4969f01" + ] + }, + "id": "5kguAb7TQZ_L", + "outputId": "aa99fbeb-3632-4cc7-b0d3-5b1bbf1c9a03" + }, + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "FIVE BERTS: PENTACHORON CONSENSUS\n", + "=================================================================\n", + " Device: cuda\n", + " Models: ['bert', 'modern', 'roberta', 'albert', 'distil']\n", + "\n", + " Loading captions (n=20000)...\n", + " Got 20000 captions\n", + "\n", + " Extracting: bert (google-bert/bert-base-uncased)...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 min_len:\n", + " captions.append(cap)\n", + " if len(captions) >= n:\n", + " break\n", + " print(f\" Got {len(captions)} captions\")\n", + " return captions\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def extract_one(model_name, short_name, captions, max_len, batch_size):\n", + " from transformers import AutoModel, AutoTokenizer\n", + " print(f\"\\n Extracting: {short_name} ({model_name})...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " dim = model.config.hidden_size\n", + " print(f\" dim={dim}, {sum(p.numel() for p in model.parameters()):,} params\")\n", + "\n", + " all_emb = []\n", + " for i in tqdm(range(0, len(captions), batch_size), desc=f\" {short_name}\"):\n", + " batch = captions[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " # Mean pool\n", + " hs = out.last_hidden_state\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (hs * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + "\n", + " emb = torch.cat(all_emb)\n", + " print(f\" Shape: {emb.shape}\")\n", + " del model\n", + " torch.cuda.empty_cache()\n", + " return emb\n", + "\n", + "\n", + "def extract_all():\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " caps_path = os.path.join(CFG.cache_dir, \"captions.json\")\n", + "\n", + " # Check cache\n", + " all_cached = all(\n", + " os.path.exists(os.path.join(CFG.cache_dir, f\"{short}.pt\"))\n", + " for _, short in MODELS)\n", + "\n", + " if all_cached:\n", + " print(\"\\n Loading cached embeddings...\")\n", + " embeds = {}\n", + " for _, short in MODELS:\n", + " embeds[short] = torch.load(\n", + " os.path.join(CFG.cache_dir, f\"{short}.pt\"), weights_only=True)\n", + " print(f\" {short}: {embeds[short].shape}\")\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " return embeds, captions\n", + "\n", + " captions = load_captions(CFG.n_samples, CFG.min_caption_len)\n", + "\n", + " embeds = {}\n", + " for model_name, short in MODELS:\n", + " emb = extract_one(model_name, short, captions,\n", + " CFG.max_len, CFG.batch_size)\n", + " # Handle dimension mismatch — ALBERT-base-v2 is 768 but some\n", + " # models might differ. Pad/truncate to 768.\n", + " if emb.shape[1] != 768:\n", + " print(f\" Adjusting {short} from {emb.shape[1]} to 768\")\n", + " if emb.shape[1] < 768:\n", + " emb = F.pad(emb, (0, 768 - emb.shape[1]))\n", + " else:\n", + " emb = emb[:, :768]\n", + " embeds[short] = emb\n", + " torch.save(emb, os.path.join(CFG.cache_dir, f\"{short}.pt\"))\n", + "\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " return embeds, captions\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# PROCRUSTES ALIGNMENT (all to first model's space)\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " \"\"\"Covariance matrix inverse square root for whitening.\"\"\"\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " \"\"\"Whitened Procrustes: normalize variance before rotating.\"\"\"\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean\n", + " Tc = T - t_mean\n", + "\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + "\n", + " N_s, D = Sc.shape\n", + "\n", + " # Whiten both distributions — every dimension contributes equally\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + "\n", + " Sc_w = Sc @ s_whiten\n", + " Tc_w = Tc @ t_whiten\n", + "\n", + " # Normalize after whitening\n", + " Sc_w = F.normalize(Sc_w, dim=-1)\n", + " Tc_w = F.normalize(Tc_w, dim=-1)\n", + "\n", + " # SVD rotation on whitened coordinates\n", + " try:\n", + " cross = Tc_w.T @ Sc_w\n", + " U, _, Vt = torch.linalg.svd(cross, full_matrices=False)\n", + " R = U @ Vt\n", + " except Exception:\n", + " R = torch.eye(D)\n", + "\n", + " # Compute unwhitener for target (to map back from whitened to target space)\n", + " t_unwhiten = torch.linalg.pinv(t_whiten)\n", + "\n", + " Sc_aligned = Sc_w @ R.T\n", + " cos_after = F.cosine_similarity(Sc_aligned, Tc_w, dim=-1).mean().item()\n", + "\n", + " return {\n", + " \"rotation\": R,\n", + " \"source_mean\": s_mean.squeeze(0),\n", + " \"target_mean\": t_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_whitener\": t_whiten,\n", + " \"target_unwhitener\": t_unwhiten,\n", + " \"cos_before\": cos_before,\n", + " \"cos_after\": cos_after,\n", + " }\n", + "\n", + "\n", + "def apply_align(emb, alignment):\n", + " \"\"\"Apply whitened Procrustes: center → whiten → rotate → unwhiten.\"\"\"\n", + " x = emb.float() - alignment[\"source_mean\"]\n", + " x = x @ alignment[\"source_whitener\"]\n", + " x = x @ alignment[\"rotation\"].T\n", + " x = x @ alignment[\"target_unwhitener\"]\n", + " return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# COMPOSITIONAL CONV5D\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def integer_compositions(n):\n", + " if n == 0: return [()]\n", + " if n == 1: return [(1,)]\n", + " result = []\n", + " for first in range(1, n + 1):\n", + " for rest in integer_compositions(n - first):\n", + " result.append((first,) + rest)\n", + " return result\n", + "\n", + "\n", + "class ConvPath(nn.Module):\n", + " def __init__(self, composition, embed_dim, hidden_dim):\n", + " super().__init__()\n", + " self.composition = composition\n", + " self.hidden_dim = hidden_dim\n", + " self.steps = nn.ModuleList()\n", + " current_dim = embed_dim\n", + " for k in composition:\n", + " self.steps.append(nn.ModuleDict({\n", + " \"proj\": nn.Linear(current_dim, hidden_dim),\n", + " \"mix\": nn.Linear(hidden_dim, hidden_dim),\n", + " \"norm\": nn.LayerNorm(hidden_dim),\n", + " }))\n", + " self.steps[-1].k_value = k\n", + " current_dim = hidden_dim\n", + " self.out = nn.Linear(hidden_dim, embed_dim)\n", + "\n", + " def forward(self, x):\n", + " B = x.shape[0]\n", + " h = x\n", + " for step in self.steps:\n", + " k = step.k_value\n", + " h = F.gelu(step[\"proj\"](h))\n", + " if k > 1 and self.hidden_dim >= k and self.hidden_dim % k == 0:\n", + " g = h.view(B, self.hidden_dim // k, k)\n", + " g = g * torch.softmax(g, dim=-1)\n", + " h = g.view(B, self.hidden_dim)\n", + " h = F.gelu(step[\"mix\"](h))\n", + " h = step[\"norm\"](h)\n", + " return self.out(h)\n", + "\n", + "\n", + "class CompConv5d(nn.Module):\n", + " def __init__(self, embed_dim, hidden_dim):\n", + " super().__init__()\n", + " self.compositions = integer_compositions(5) # 16 paths\n", + " self.paths = nn.ModuleList([\n", + " ConvPath(c, embed_dim, hidden_dim) for c in self.compositions])\n", + " self.weights = nn.Parameter(torch.ones(16) / 16)\n", + " self.fusion = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim), nn.GELU(), nn.LayerNorm(embed_dim))\n", + "\n", + " def forward(self, x):\n", + " outs = torch.stack([p(x) for p in self.paths], dim=1)\n", + " w = F.softmax(self.weights, dim=0)\n", + " return self.fusion((outs * w.view(1, -1, 1)).sum(1))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# FIVE-ANCHOR MODEL\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class FiveAnchorModel(nn.Module):\n", + " def __init__(self, embed_dim=768, hidden_dim=256, n_anchors=5):\n", + " super().__init__()\n", + " self.n_anchors = n_anchors\n", + " self.conv5 = CompConv5d(embed_dim, hidden_dim)\n", + "\n", + " # Combine all 5 anchors + structural signal\n", + " self.combine = nn.Sequential(\n", + " nn.Linear(embed_dim * (n_anchors + 1), embed_dim),\n", + " nn.GELU(), nn.LayerNorm(embed_dim))\n", + " self.output = nn.Sequential(\n", + " nn.Linear(embed_dim, embed_dim),\n", + " nn.GELU(), nn.Linear(embed_dim, embed_dim))\n", + "\n", + " def forward(self, anchors):\n", + " \"\"\"\n", + " anchors: list of 5 tensors, each (B, 768)\n", + " Returns: (B, 768) unified embedding\n", + " \"\"\"\n", + " # Structural signal: sum of all pairwise differences\n", + " # This captures the full 5-way structural relationship\n", + " structural = torch.zeros_like(anchors[0])\n", + " for i in range(self.n_anchors):\n", + " for j in range(i+1, self.n_anchors):\n", + " structural = structural + (anchors[i] - anchors[j])\n", + "\n", + " # Decompose through conv5d\n", + " decomposed = self.conv5(structural)\n", + "\n", + " # Mean of all anchors\n", + " mean_anchor = sum(anchors) / self.n_anchors\n", + "\n", + " # Combine: decomposed + all anchors\n", + " combined = self.combine(torch.cat([decomposed] + anchors, dim=-1))\n", + "\n", + " return F.normalize(self.output(combined + mean_anchor), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# LOSSES\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "\n", + "def cv_loss(emb, target=0.20, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "def svd_align(a, b):\n", + " A = F.normalize(a.float(), dim=-1)\n", + " B_e = F.normalize(b.float(), dim=-1)\n", + " A = A - A.mean(0, keepdim=True)\n", + " B_e = B_e - B_e.mean(0, keepdim=True)\n", + " N, D = A.shape\n", + " try:\n", + " S = torch.linalg.svdvals(A @ B_e.T) if N < D else torch.linalg.svdvals(A.T @ B_e)\n", + " except Exception:\n", + " return torch.tensor(0.0, device=a.device, requires_grad=True)\n", + " return 1.0 - S.sum() / (math.sqrt(N) * D)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def train():\n", + " # ── Seed everything ──\n", + " torch.manual_seed(CFG.seed)\n", + " torch.cuda.manual_seed_all(CFG.seed)\n", + " np.random.seed(CFG.seed)\n", + " print(f\"\\n Seed: {CFG.seed}\")\n", + "\n", + " # ── Extract ──\n", + " embeds, captions = extract_all()\n", + "\n", + " # ── Align all to first model's space ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PROCRUSTES ALIGNMENT (all → bert space)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref_name = MODELS[0][1] # bert\n", + " aligned = {}\n", + " align_info = {}\n", + "\n", + " for _, short in MODELS:\n", + " # All models go through whitened Procrustes — including reference\n", + " # Self-alignment whitens the reference into the same normalized space\n", + " info = procrustes_align(embeds[short], embeds[ref_name])\n", + " aligned[short] = apply_align(embeds[short], info)\n", + " align_info[short] = info\n", + " label = \" (reference)\" if short == ref_name else \"\"\n", + " print(f\" {short:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}{label}\")\n", + "\n", + " # Pairwise cosines after alignment\n", + " print(f\"\\n Pairwise cosines (post-alignment):\")\n", + " names = [s for _, s in MODELS]\n", + " for i in range(len(names)):\n", + " for j in range(i+1, len(names)):\n", + " cos = F.cosine_similarity(\n", + " aligned[names[i]][:5000], aligned[names[j]][:5000], dim=-1\n", + " ).mean().item()\n", + " print(f\" {names[i]:8s} ↔ {names[j]:8s}: {cos:.4f}\")\n", + "\n", + " # Split\n", + " n_val = 2000\n", + " n_train = CFG.n_samples - n_val\n", + " train_data = {k: v[:n_train].to(DEVICE) for k, v in aligned.items()}\n", + " val_data = {k: v[n_train:].to(DEVICE) for k, v in aligned.items()}\n", + " anchor_names = [s for _, s in MODELS]\n", + " print(f\"\\n Train: {n_train}, Val: {n_val}\")\n", + "\n", + " # ── Model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"MODEL: FiveAnchorModel + CompConv5d\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = FiveAnchorModel(768, CFG.hidden_dim, 5).to(DEVICE)\n", + " n_params = sum(p.numel() for p in model.parameters())\n", + " print(f\" Parameters: {n_params:,}\")\n", + " print(f\" Conv5d paths: {len(model.conv5.compositions)} (2^4 = 16)\")\n", + "\n", + " # ── Train ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({CFG.epochs} epochs)\")\n", + " print(f\" 5 InfoNCE + 5 SVD + 1 CV = 11 losses\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " optimizer = torch.optim.AdamW(model.parameters(), lr=CFG.lr,\n", + " weight_decay=CFG.weight_decay)\n", + " n_batches = n_train // CFG.train_batch\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * CFG.epochs, eta_min=1e-6)\n", + "\n", + " all_metrics = {\"seed\": CFG.seed, \"epochs\": [], \"path_weights\": [], \"alignment\": align_info}\n", + "\n", + " for epoch in range(CFG.epochs):\n", + " model.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " accs = {n: 0 for n in anchor_names}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, CFG.train_batch):\n", + " idx = perm[i:i+CFG.train_batch]\n", + " if len(idx) < 8: continue\n", + "\n", + " batch_anchors = [train_data[name][idx] for name in anchor_names]\n", + " emb = model(batch_anchors)\n", + "\n", + " # 5 InfoNCE losses + 5 SVD losses\n", + " loss = torch.tensor(0.0, device=DEVICE)\n", + " for k, name in enumerate(anchor_names):\n", + " l_nce, acc = infonce(emb, batch_anchors[k])\n", + " l_svd = svd_align(emb, batch_anchors[k])\n", + " loss = loss + CFG.nce_weight * l_nce + CFG.svd_weight * l_svd\n", + " accs[name] += acc\n", + "\n", + " # CV loss\n", + " l_cv = cv_loss(emb)\n", + " loss = loss + CFG.cv_weight * l_cv\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(model.parameters(), CFG.grad_clip)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " total_loss += loss.item()\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Val\n", + " model.eval()\n", + " with torch.no_grad():\n", + " val_anchors = [val_data[name] for name in anchor_names]\n", + " val_emb = model(val_anchors)\n", + " val_accs = {}\n", + " for k, name in enumerate(anchor_names):\n", + " _, va = infonce(val_emb, val_anchors[k])\n", + " val_accs[name] = va\n", + " val_cv = cv_metric(val_emb)\n", + "\n", + " # Path weights\n", + " pw = F.softmax(model.conv5.weights, dim=0).detach().cpu().tolist()\n", + "\n", + " # Top 3 paths\n", + " sorted_paths = sorted(zip(model.conv5.compositions, pw), key=lambda x: -x[1])\n", + " top3 = \" \".join(f\"{str(c)}={w:.4f}\" for c, w in sorted_paths[:3])\n", + "\n", + " # Acc summary\n", + " acc_str = \"/\".join(f\"{accs[name]/d:.3f}\" for name in anchor_names)\n", + " vacc_str = \"/\".join(f\"{val_accs[name]:.3f}\" for name in anchor_names)\n", + "\n", + " summary = {\n", + " \"epoch\": epoch + 1, \"loss\": total_loss / d,\n", + " \"train_accs\": {n: accs[n]/d for n in anchor_names},\n", + " \"val_accs\": val_accs, \"val_cv\": val_cv,\n", + " }\n", + " all_metrics[\"epochs\"].append(summary)\n", + " all_metrics[\"path_weights\"].append(pw)\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} \"\n", + " f\"acc={acc_str} val={vacc_str} cv={val_cv:.3f} top: {top3}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # FINAL ANALYSIS\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FINAL PATH WEIGHT ANALYSIS (16 paths)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " pw = F.softmax(model.conv5.weights, dim=0).detach().cpu()\n", + " labels = {1: \"indep\", 2: \"pair\", 3: \"3-way\", 4: \"4-way\", 5: \"5-way\"}\n", + " for comp, w in sorted(zip(model.conv5.compositions, pw.tolist()), key=lambda x: -x[1]):\n", + " bar = \"█\" * int(w * 80)\n", + " interp = \" → \".join(labels.get(k, f\"{k}\") for k in comp)\n", + " print(f\" {str(comp):<25} {w:.4f} {bar} ({interp})\")\n", + "\n", + " # Weight spread\n", + " pw_arr = pw.numpy()\n", + " print(f\"\\n Weight spread: min={pw_arr.min():.4f} max={pw_arr.max():.4f} \"\n", + " f\"std={pw_arr.std():.6f} range={pw_arr.max()-pw_arr.min():.6f}\")\n", + " print(f\" Uniform would be: {1/16:.4f} = 0.0625\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"CONSENSUS GEOMETRY\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model.eval()\n", + " with torch.no_grad():\n", + " val_anchors = [val_data[name] for name in anchor_names]\n", + " val_emb = model(val_anchors).cpu()\n", + "\n", + " # Cosine to each anchor\n", + " for name in anchor_names:\n", + " cos = F.cosine_similarity(\n", + " val_emb, val_data[name].cpu(), dim=-1).mean().item()\n", + " print(f\" cos(output, {name:10s}): {cos:.4f}\")\n", + "\n", + " # Cosine to geometric centroid\n", + " centroid = F.normalize(sum(val_data[n].cpu() for n in anchor_names) / 5, dim=-1)\n", + " cos_cent = F.cosine_similarity(val_emb, centroid, dim=-1).mean().item()\n", + " print(f\" cos(output, centroid): {cos_cent:.4f}\")\n", + "\n", + " # CV comparison\n", + " out_cv = cv_metric(val_emb[:1000])\n", + " print(f\"\\n CV output: {out_cv:.4f}\")\n", + " for name in anchor_names:\n", + " acv = cv_metric(val_data[name].cpu()[:1000])\n", + " print(f\" CV {name:10s}: {acv:.4f}\")\n", + "\n", + " # Pairwise cosines between output and each anchor\n", + " print(f\"\\n Equidistance check (should be ~equal):\")\n", + " cosines = []\n", + " for name in anchor_names:\n", + " cos = F.cosine_similarity(\n", + " val_emb, val_data[name].cpu(), dim=-1).mean().item()\n", + " cosines.append(cos)\n", + " print(f\" Range: {max(cosines)-min(cosines):.6f}\")\n", + " print(f\" Std: {np.std(cosines):.6f}\")\n", + "\n", + " # Save\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " with open(os.path.join(CFG.cache_dir, f\"five_results_seed{CFG.seed}.json\"), \"w\") as f:\n", + " json.dump(all_metrics, f, indent=2, default=str)\n", + " torch.save(model.state_dict(), os.path.join(CFG.cache_dir, f\"five_model_seed{CFG.seed}.pt\"))\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " # Default: run 3 seeds\n", + " seeds = [42, 43, 44, 12341, 12323, 8675309]\n", + "\n", + " for seed in seeds:\n", + " CFG.seed = seed\n", + " train()\n", + " print(f\"\\n{'#'*65}\")\n", + " print(f\"# SEED {seed} COMPLETE\")\n", + " print(f\"{'#'*65}\\n\")\n", + " torch.cuda.empty_cache()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "tzbRxCfQTbqQ", + "outputId": "68cd9008-a3c8-4c85-d11f-a5271d056499" + }, + "execution_count": 6, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "FIVE BERTS: PENTACHORON CONSENSUS\n", + "=================================================================\n", + " Device: cuda\n", + " Models: ['bert', 'modern', 'roberta', 'albert', 'distil']\n", + "\n", + " Seed: 42\n", + "\n", + " Loading cached embeddings...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + "\n", + "=================================================================\n", + "PROCRUSTES ALIGNMENT (all → bert space)\n", + "=================================================================\n", + " bert : cos 1.0000 → 1.0000 (reference)\n", + " modern : cos -0.0025 → 0.4849\n", + " roberta : cos -0.0037 → 0.5138\n", + " albert : cos -0.0004 → 0.4888\n", + " distil : cos 0.8567 → 0.6557\n", + "\n", + " Pairwise cosines (post-alignment):\n", + " bert ↔ modern : 0.8357\n", + " bert ↔ roberta : 0.8685\n", + " bert ↔ albert : 0.8413\n", + " bert ↔ distil : 0.9314\n", + " modern ↔ roberta : 0.8040\n", + " modern ↔ albert : 0.7777\n", + " modern ↔ distil : 0.8224\n", + " roberta ↔ albert : 0.8039\n", + " roberta ↔ distil : 0.8528\n", + " albert ↔ distil : 0.8269\n", + "\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "MODEL: FiveAnchorModel + CompConv5d\n", + "=================================================================\n", + " Parameters: 16,910,352\n", + " Conv5d paths: 16 (2^4 = 16)\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " 5 InfoNCE + 5 SVD + 1 CV = 11 losses\n", + "=================================================================\n", + " E 1: 4s loss=3.4940 acc=0.927/0.912/0.920/0.920/0.924 val=1.000/1.000/1.000/0.999/1.000 cv=0.119 top: (2, 1, 2)=0.0628 (2, 1, 1, 1)=0.0628 (4, 1)=0.0627\n", + " E 2: 4s loss=0.5318 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.105 top: (2, 1, 2)=0.0628 (2, 1, 1, 1)=0.0628 (4, 1)=0.0628\n", + " E 3: 4s loss=0.4549 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.105 top: (2, 1, 2)=0.0629 (2, 1, 1, 1)=0.0628 (4, 1)=0.0628\n", + " E 4: 4s loss=0.4216 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.104 top: (2, 1, 2)=0.0629 (4, 1)=0.0628 (2, 1, 1, 1)=0.0628\n", + " E 5: 4s loss=0.4071 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.098 top: (2, 1, 2)=0.0629 (4, 1)=0.0628 (2, 1, 1, 1)=0.0628\n", + " E 6: 4s loss=0.3956 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.100 top: (2, 1, 2)=0.0629 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E 7: 4s loss=0.3864 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.097 top: (2, 1, 2)=0.0628 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E 8: 4s loss=0.3797 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.092 top: (2, 1, 2)=0.0628 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E 9: 4s loss=0.3735 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.096 top: (2, 1, 2)=0.0628 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E10: 4s loss=0.3690 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.094 top: (2, 1, 2)=0.0628 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E11: 4s loss=0.3653 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.090 top: (2, 1, 2)=0.0628 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E12: 4s loss=0.3643 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (2, 1, 2)=0.0628 (4, 1)=0.0628 (2, 1, 1, 1)=0.0627\n", + " E13: 4s loss=0.3601 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.090 top: (2, 1, 2)=0.0628 (4, 1)=0.0627 (2, 2, 1)=0.0626\n", + " E14: 4s loss=0.3583 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.089 top: (2, 1, 2)=0.0628 (4, 1)=0.0627 (2, 2, 1)=0.0626\n", + " E15: 4s loss=0.3545 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.081 top: (2, 1, 2)=0.0627 (4, 1)=0.0627 (2, 2, 1)=0.0626\n", + " E16: 4s loss=0.3527 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.078 top: (2, 1, 2)=0.0627 (4, 1)=0.0627 (2, 2, 1)=0.0626\n", + " E17: 4s loss=0.3519 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.079 top: (4, 1)=0.0627 (2, 1, 2)=0.0627 (2, 2, 1)=0.0626\n", + " E18: 4s loss=0.3522 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.085 top: (4, 1)=0.0627 (2, 1, 2)=0.0627 (2, 2, 1)=0.0626\n", + " E19: 4s loss=0.3505 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.072 top: (4, 1)=0.0627 (2, 1, 2)=0.0627 (2, 2, 1)=0.0626\n", + " E20: 4s loss=0.3505 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.078 top: (4, 1)=0.0627 (2, 1, 2)=0.0627 (2, 2, 1)=0.0626\n", + "\n", + "=================================================================\n", + "FINAL PATH WEIGHT ANALYSIS (16 paths)\n", + "=================================================================\n", + " (4, 1) 0.0627 █████ (4-way → indep)\n", + " (2, 1, 2) 0.0627 █████ (pair → indep → pair)\n", + " (2, 2, 1) 0.0626 █████ (pair → pair → indep)\n", + " (2, 1, 1, 1) 0.0626 █████ (pair → indep → indep → indep)\n", + " (1, 2, 2) 0.0626 █████ (indep → pair → pair)\n", + " (2, 3) 0.0626 █████ (pair → 3-way)\n", + " (1, 1, 3) 0.0626 █████ (indep → indep → 3-way)\n", + " (1, 2, 1, 1) 0.0625 █████ (indep → pair → indep → indep)\n", + " (1, 1, 2, 1) 0.0625 ████ (indep → indep → pair → indep)\n", + " (1, 1, 1, 1, 1) 0.0624 ████ (indep → indep → indep → indep → indep)\n", + " (1, 3, 1) 0.0624 ████ (indep → 3-way → indep)\n", + " (1, 1, 1, 2) 0.0624 ████ (indep → indep → indep → pair)\n", + " (1, 4) 0.0624 ████ (indep → 4-way)\n", + " (5,) 0.0624 ████ (5-way)\n", + " (3, 1, 1) 0.0623 ████ (3-way → indep → indep)\n", + " (3, 2) 0.0622 ████ (3-way → pair)\n", + "\n", + " Weight spread: min=0.0622 max=0.0627 std=0.000149 range=0.000586\n", + " Uniform would be: 0.0625 = 0.0625\n", + "\n", + "=================================================================\n", + "CONSENSUS GEOMETRY\n", + "=================================================================\n", + " cos(output, bert ): 0.8420\n", + " cos(output, modern ): 0.8025\n", + " cos(output, roberta ): 0.8192\n", + " cos(output, albert ): 0.8026\n", + " cos(output, distil ): 0.8376\n", + " cos(output, centroid): 0.9088\n", + "\n", + " CV output: 0.0810\n", + " CV bert : 0.3649\n", + " CV modern : 0.3245\n", + " CV roberta : 0.3559\n", + " CV albert : 0.3863\n", + " CV distil : 0.4015\n", + "\n", + " Equidistance check (should be ~equal):\n", + " Range: 0.039510\n", + " Std: 0.016736\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n", + "\n", + "#################################################################\n", + "# SEED 42 COMPLETE\n", + "#################################################################\n", + "\n", + "\n", + " Seed: 43\n", + "\n", + " Loading cached embeddings...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + "\n", + "=================================================================\n", + "PROCRUSTES ALIGNMENT (all → bert space)\n", + "=================================================================\n", + " bert : cos 1.0000 → 1.0000 (reference)\n", + " modern : cos -0.0025 → 0.4849\n", + " roberta : cos -0.0037 → 0.5138\n", + " albert : cos -0.0004 → 0.4888\n", + " distil : cos 0.8567 → 0.6557\n", + "\n", + " Pairwise cosines (post-alignment):\n", + " bert ↔ modern : 0.8357\n", + " bert ↔ roberta : 0.8685\n", + " bert ↔ albert : 0.8413\n", + " bert ↔ distil : 0.9314\n", + " modern ↔ roberta : 0.8040\n", + " modern ↔ albert : 0.7777\n", + " modern ↔ distil : 0.8224\n", + " roberta ↔ albert : 0.8039\n", + " roberta ↔ distil : 0.8528\n", + " albert ↔ distil : 0.8269\n", + "\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "MODEL: FiveAnchorModel + CompConv5d\n", + "=================================================================\n", + " Parameters: 16,910,352\n", + " Conv5d paths: 16 (2^4 = 16)\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " 5 InfoNCE + 5 SVD + 1 CV = 11 losses\n", + "=================================================================\n", + " E 1: 4s loss=3.4673 acc=0.929/0.916/0.922/0.920/0.925 val=1.000/0.998/1.000/0.997/1.000 cv=0.120 top: (1, 2, 2)=0.0628 (4, 1)=0.0627 (1, 1, 1, 1, 1)=0.0627\n", + " E 2: 4s loss=0.5333 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.116 top: (1, 2, 2)=0.0628 (4, 1)=0.0627 (1, 1, 1, 1, 1)=0.0627\n", + " E 3: 4s loss=0.4559 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.114 top: (1, 2, 2)=0.0628 (4, 1)=0.0627 (1, 1, 1, 1, 1)=0.0627\n", + " E 4: 4s loss=0.4218 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.106 top: (1, 2, 2)=0.0628 (4, 1)=0.0627 (1, 1, 1, 1, 1)=0.0627\n", + " E 5: 4s loss=0.4062 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.100 top: (1, 2, 2)=0.0628 (4, 1)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E 6: 4s loss=0.3937 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.100 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E 7: 4s loss=0.3849 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.103 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E 8: 4s loss=0.3791 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.099 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E 9: 4s loss=0.3744 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.098 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E10: 4s loss=0.3697 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E11: 4s loss=0.3647 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.085 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E12: 4s loss=0.3607 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0627\n", + " E13: 4s loss=0.3578 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.074 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E14: 4s loss=0.3572 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.080 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E15: 4s loss=0.3550 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.083 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E16: 4s loss=0.3544 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E17: 4s loss=0.3513 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.077 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E18: 4s loss=0.3507 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E19: 4s loss=0.3496 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + " E20: 4s loss=0.3505 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (4, 1)=0.0627 (1, 2, 2)=0.0627 (1, 2, 1, 1)=0.0626\n", + "\n", + "=================================================================\n", + "FINAL PATH WEIGHT ANALYSIS (16 paths)\n", + "=================================================================\n", + " (4, 1) 0.0627 █████ (4-way → indep)\n", + " (1, 2, 2) 0.0627 █████ (indep → pair → pair)\n", + " (1, 2, 1, 1) 0.0626 █████ (indep → pair → indep → indep)\n", + " (2, 3) 0.0626 █████ (pair → 3-way)\n", + " (1, 4) 0.0626 █████ (indep → 4-way)\n", + " (2, 1, 2) 0.0626 █████ (pair → indep → pair)\n", + " (1, 1, 1, 1, 1) 0.0626 █████ (indep → indep → indep → indep → indep)\n", + " (2, 2, 1) 0.0626 █████ (pair → pair → indep)\n", + " (3, 1, 1) 0.0625 █████ (3-way → indep → indep)\n", + " (3, 2) 0.0625 █████ (3-way → pair)\n", + " (2, 1, 1, 1) 0.0625 ████ (pair → indep → indep → indep)\n", + " (1, 1, 2, 1) 0.0624 ████ (indep → indep → pair → indep)\n", + " (1, 3, 1) 0.0623 ████ (indep → 3-way → indep)\n", + " (1, 1, 3) 0.0623 ████ (indep → indep → 3-way)\n", + " (5,) 0.0623 ████ (5-way)\n", + " (1, 1, 1, 2) 0.0623 ████ (indep → indep → indep → pair)\n", + "\n", + " Weight spread: min=0.0623 max=0.0627 std=0.000151 range=0.000448\n", + " Uniform would be: 0.0625 = 0.0625\n", + "\n", + "=================================================================\n", + "CONSENSUS GEOMETRY\n", + "=================================================================\n", + " cos(output, bert ): 0.8421\n", + " cos(output, modern ): 0.8024\n", + " cos(output, roberta ): 0.8192\n", + " cos(output, albert ): 0.8027\n", + " cos(output, distil ): 0.8377\n", + " cos(output, centroid): 0.9089\n", + "\n", + " CV output: 0.0834\n", + " CV bert : 0.3983\n", + " CV modern : 0.3244\n", + " CV roberta : 0.3226\n", + " CV albert : 0.4035\n", + " CV distil : 0.4327\n", + "\n", + " Equidistance check (should be ~equal):\n", + " Range: 0.039687\n", + " Std: 0.016773\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n", + "\n", + "#################################################################\n", + "# SEED 43 COMPLETE\n", + "#################################################################\n", + "\n", + "\n", + " Seed: 44\n", + "\n", + " Loading cached embeddings...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + "\n", + "=================================================================\n", + "PROCRUSTES ALIGNMENT (all → bert space)\n", + "=================================================================\n", + " bert : cos 1.0000 → 1.0000 (reference)\n", + " modern : cos -0.0025 → 0.4849\n", + " roberta : cos -0.0037 → 0.5138\n", + " albert : cos -0.0004 → 0.4888\n", + " distil : cos 0.8567 → 0.6557\n", + "\n", + " Pairwise cosines (post-alignment):\n", + " bert ↔ modern : 0.8357\n", + " bert ↔ roberta : 0.8685\n", + " bert ↔ albert : 0.8413\n", + " bert ↔ distil : 0.9314\n", + " modern ↔ roberta : 0.8040\n", + " modern ↔ albert : 0.7777\n", + " modern ↔ distil : 0.8224\n", + " roberta ↔ albert : 0.8039\n", + " roberta ↔ distil : 0.8528\n", + " albert ↔ distil : 0.8269\n", + "\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "MODEL: FiveAnchorModel + CompConv5d\n", + "=================================================================\n", + " Parameters: 16,910,352\n", + " Conv5d paths: 16 (2^4 = 16)\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " 5 InfoNCE + 5 SVD + 1 CV = 11 losses\n", + "=================================================================\n", + " E 1: 4s loss=3.5689 acc=0.922/0.908/0.916/0.913/0.920 val=1.000/0.998/1.000/0.998/1.000 cv=0.121 top: (2, 1, 1, 1)=0.0627 (1, 1, 2, 1)=0.0627 (4, 1)=0.0627\n", + " E 2: 4s loss=0.5315 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.109 top: (2, 1, 1, 1)=0.0628 (1, 1, 2, 1)=0.0627 (4, 1)=0.0627\n", + " E 3: 4s loss=0.4524 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.103 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (1, 1, 2, 1)=0.0627\n", + " E 4: 4s loss=0.4229 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.100 top: (2, 1, 1, 1)=0.0629 (4, 1)=0.0627 (1, 1, 2, 1)=0.0627\n", + " E 5: 4s loss=0.4060 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.093 top: (2, 1, 1, 1)=0.0629 (4, 1)=0.0627 (1, 1, 2, 1)=0.0627\n", + " E 6: 4s loss=0.3944 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.093 top: (2, 1, 1, 1)=0.0629 (4, 1)=0.0627 (1, 1, 2, 1)=0.0627\n", + " E 7: 4s loss=0.3835 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.090 top: (2, 1, 1, 1)=0.0629 (4, 1)=0.0627 (2, 1, 2)=0.0627\n", + " E 8: 4s loss=0.3782 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.086 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 1, 2)=0.0627\n", + " E 9: 4s loss=0.3759 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 1, 2)=0.0626\n", + " E10: 4s loss=0.3679 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.099 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E11: 4s loss=0.3638 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E12: 4s loss=0.3616 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E13: 4s loss=0.3597 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.080 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E14: 4s loss=0.3586 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.078 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E15: 4s loss=0.3550 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.088 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E16: 4s loss=0.3537 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.086 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E17: 4s loss=0.3530 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.081 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0627\n", + " E18: 4s loss=0.3518 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.087 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E19: 4s loss=0.3499 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.089 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + " E20: 4s loss=0.3488 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.089 top: (2, 1, 1, 1)=0.0628 (4, 1)=0.0627 (2, 3)=0.0626\n", + "\n", + "=================================================================\n", + "FINAL PATH WEIGHT ANALYSIS (16 paths)\n", + "=================================================================\n", + " (2, 1, 1, 1) 0.0628 █████ (pair → indep → indep → indep)\n", + " (4, 1) 0.0627 █████ (4-way → indep)\n", + " (2, 3) 0.0626 █████ (pair → 3-way)\n", + " (1, 1, 3) 0.0626 █████ (indep → indep → 3-way)\n", + " (2, 1, 2) 0.0626 █████ (pair → indep → pair)\n", + " (1, 1, 2, 1) 0.0626 █████ (indep → indep → pair → indep)\n", + " (2, 2, 1) 0.0625 █████ (pair → pair → indep)\n", + " (1, 4) 0.0625 █████ (indep → 4-way)\n", + " (1, 2, 1, 1) 0.0625 ████ (indep → pair → indep → indep)\n", + " (1, 3, 1) 0.0625 ████ (indep → 3-way → indep)\n", + " (1, 1, 1, 1, 1) 0.0624 ████ (indep → indep → indep → indep → indep)\n", + " (1, 1, 1, 2) 0.0624 ████ (indep → indep → indep → pair)\n", + " (3, 1, 1) 0.0624 ████ (3-way → indep → indep)\n", + " (5,) 0.0623 ████ (5-way)\n", + " (1, 2, 2) 0.0623 ████ (indep → pair → pair)\n", + " (3, 2) 0.0623 ████ (3-way → pair)\n", + "\n", + " Weight spread: min=0.0623 max=0.0628 std=0.000133 range=0.000484\n", + " Uniform would be: 0.0625 = 0.0625\n", + "\n", + "=================================================================\n", + "CONSENSUS GEOMETRY\n", + "=================================================================\n", + " cos(output, bert ): 0.8422\n", + " cos(output, modern ): 0.8026\n", + " cos(output, roberta ): 0.8192\n", + " cos(output, albert ): 0.8029\n", + " cos(output, distil ): 0.8377\n", + " cos(output, centroid): 0.9089\n", + "\n", + " CV output: 0.0913\n", + " CV bert : 0.4204\n", + " CV modern : 0.3488\n", + " CV roberta : 0.3288\n", + " CV albert : 0.3748\n", + " CV distil : 0.3881\n", + "\n", + " Equidistance check (should be ~equal):\n", + " Range: 0.039557\n", + " Std: 0.016713\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n", + "\n", + "#################################################################\n", + "# SEED 44 COMPLETE\n", + "#################################################################\n", + "\n", + "\n", + " Seed: 12341\n", + "\n", + " Loading cached embeddings...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + "\n", + "=================================================================\n", + "PROCRUSTES ALIGNMENT (all → bert space)\n", + "=================================================================\n", + " bert : cos 1.0000 → 1.0000 (reference)\n", + " modern : cos -0.0025 → 0.4849\n", + " roberta : cos -0.0037 → 0.5138\n", + " albert : cos -0.0004 → 0.4888\n", + " distil : cos 0.8567 → 0.6557\n", + "\n", + " Pairwise cosines (post-alignment):\n", + " bert ↔ modern : 0.8357\n", + " bert ↔ roberta : 0.8685\n", + " bert ↔ albert : 0.8413\n", + " bert ↔ distil : 0.9314\n", + " modern ↔ roberta : 0.8040\n", + " modern ↔ albert : 0.7777\n", + " modern ↔ distil : 0.8224\n", + " roberta ↔ albert : 0.8039\n", + " roberta ↔ distil : 0.8528\n", + " albert ↔ distil : 0.8269\n", + "\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "MODEL: FiveAnchorModel + CompConv5d\n", + "=================================================================\n", + " Parameters: 16,910,352\n", + " Conv5d paths: 16 (2^4 = 16)\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " 5 InfoNCE + 5 SVD + 1 CV = 11 losses\n", + "=================================================================\n", + " E 1: 4s loss=3.6163 acc=0.924/0.908/0.915/0.914/0.920 val=1.000/0.999/1.000/1.000/1.000 cv=0.121 top: (2, 1, 1, 1)=0.0628 (2, 2, 1)=0.0627 (2, 3)=0.0627\n", + " E 2: 4s loss=0.5378 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.114 top: (2, 1, 1, 1)=0.0628 (2, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E 3: 4s loss=0.4537 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.103 top: (2, 1, 1, 1)=0.0629 (2, 2, 1)=0.0628 (2, 3)=0.0628\n", + " E 4: 4s loss=0.4206 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.106 top: (2, 1, 1, 1)=0.0628 (2, 3)=0.0628 (2, 2, 1)=0.0628\n", + " E 5: 4s loss=0.4046 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.086 top: (2, 1, 1, 1)=0.0628 (2, 3)=0.0628 (2, 2, 1)=0.0628\n", + " E 6: 4s loss=0.3930 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.100 top: (2, 3)=0.0628 (2, 1, 1, 1)=0.0628 (2, 2, 1)=0.0628\n", + " E 7: 4s loss=0.3839 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.092 top: (2, 3)=0.0628 (2, 1, 1, 1)=0.0628 (2, 2, 1)=0.0628\n", + " E 8: 4s loss=0.3787 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.092 top: (2, 3)=0.0628 (2, 1, 1, 1)=0.0628 (2, 2, 1)=0.0627\n", + " E 9: 4s loss=0.3735 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.080 top: (2, 3)=0.0628 (2, 1, 1, 1)=0.0627 (2, 2, 1)=0.0627\n", + " E10: 4s loss=0.3689 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.083 top: (2, 3)=0.0628 (2, 1, 1, 1)=0.0627 (2, 2, 1)=0.0627\n", + " E11: 4s loss=0.3650 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.088 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (2, 1, 1, 1)=0.0627\n", + " E12: 4s loss=0.3623 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.080 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (2, 1, 1, 1)=0.0627\n", + " E13: 4s loss=0.3591 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (2, 1, 1, 1)=0.0627\n", + " E14: 4s loss=0.3565 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (2, 1, 1, 1)=0.0627\n", + " E15: 4s loss=0.3558 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (2, 1, 1, 1)=0.0626\n", + " E16: 4s loss=0.3537 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.081 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (2, 1, 1, 1)=0.0626\n", + " E17: 4s loss=0.3532 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (4, 1)=0.0626\n", + " E18: 4s loss=0.3509 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (4, 1)=0.0626\n", + " E19: 4s loss=0.3502 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.071 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (4, 1)=0.0626\n", + " E20: 4s loss=0.3505 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.089 top: (2, 3)=0.0628 (2, 2, 1)=0.0627 (4, 1)=0.0626\n", + "\n", + "=================================================================\n", + "FINAL PATH WEIGHT ANALYSIS (16 paths)\n", + "=================================================================\n", + " (2, 3) 0.0628 █████ (pair → 3-way)\n", + " (2, 2, 1) 0.0627 █████ (pair → pair → indep)\n", + " (4, 1) 0.0626 █████ (4-way → indep)\n", + " (2, 1, 1, 1) 0.0626 █████ (pair → indep → indep → indep)\n", + " (1, 1, 2, 1) 0.0626 █████ (indep → indep → pair → indep)\n", + " (1, 4) 0.0625 █████ (indep → 4-way)\n", + " (1, 2, 2) 0.0625 █████ (indep → pair → pair)\n", + " (2, 1, 2) 0.0625 █████ (pair → indep → pair)\n", + " (3, 1, 1) 0.0625 ████ (3-way → indep → indep)\n", + " (1, 1, 3) 0.0625 ████ (indep → indep → 3-way)\n", + " (1, 3, 1) 0.0624 ████ (indep → 3-way → indep)\n", + " (1, 2, 1, 1) 0.0624 ████ (indep → pair → indep → indep)\n", + " (5,) 0.0624 ████ (5-way)\n", + " (1, 1, 1, 2) 0.0623 ████ (indep → indep → indep → pair)\n", + " (3, 2) 0.0623 ████ (3-way → pair)\n", + " (1, 1, 1, 1, 1) 0.0623 ████ (indep → indep → indep → indep → indep)\n", + "\n", + " Weight spread: min=0.0623 max=0.0628 std=0.000140 range=0.000531\n", + " Uniform would be: 0.0625 = 0.0625\n", + "\n", + "=================================================================\n", + "CONSENSUS GEOMETRY\n", + "=================================================================\n", + " cos(output, bert ): 0.8419\n", + " cos(output, modern ): 0.8026\n", + " cos(output, roberta ): 0.8192\n", + " cos(output, albert ): 0.8028\n", + " cos(output, distil ): 0.8377\n", + " cos(output, centroid): 0.9088\n", + "\n", + " CV output: 0.0791\n", + " CV bert : 0.3944\n", + " CV modern : 0.3594\n", + " CV roberta : 0.3726\n", + " CV albert : 0.3543\n", + " CV distil : 0.4075\n", + "\n", + " Equidistance check (should be ~equal):\n", + " Range: 0.039320\n", + " Std: 0.016674\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n", + "\n", + "#################################################################\n", + "# SEED 12341 COMPLETE\n", + "#################################################################\n", + "\n", + "\n", + " Seed: 12323\n", + "\n", + " Loading cached embeddings...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + "\n", + "=================================================================\n", + "PROCRUSTES ALIGNMENT (all → bert space)\n", + "=================================================================\n", + " bert : cos 1.0000 → 1.0000 (reference)\n", + " modern : cos -0.0025 → 0.4849\n", + " roberta : cos -0.0037 → 0.5138\n", + " albert : cos -0.0004 → 0.4888\n", + " distil : cos 0.8567 → 0.6557\n", + "\n", + " Pairwise cosines (post-alignment):\n", + " bert ↔ modern : 0.8357\n", + " bert ↔ roberta : 0.8685\n", + " bert ↔ albert : 0.8413\n", + " bert ↔ distil : 0.9314\n", + " modern ↔ roberta : 0.8040\n", + " modern ↔ albert : 0.7777\n", + " modern ↔ distil : 0.8224\n", + " roberta ↔ albert : 0.8039\n", + " roberta ↔ distil : 0.8528\n", + " albert ↔ distil : 0.8269\n", + "\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "MODEL: FiveAnchorModel + CompConv5d\n", + "=================================================================\n", + " Parameters: 16,910,352\n", + " Conv5d paths: 16 (2^4 = 16)\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " 5 InfoNCE + 5 SVD + 1 CV = 11 losses\n", + "=================================================================\n", + " E 1: 4s loss=3.6015 acc=0.924/0.908/0.916/0.913/0.919 val=1.000/0.998/1.000/0.999/1.000 cv=0.130 top: (1, 1, 1, 2)=0.0627 (3, 1, 1)=0.0627 (4, 1)=0.0627\n", + " E 2: 4s loss=0.5371 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/0.999/1.000/0.999/1.000 cv=0.105 top: (1, 1, 1, 2)=0.0628 (3, 1, 1)=0.0628 (4, 1)=0.0627\n", + " E 3: 4s loss=0.4539 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.104 top: (1, 1, 1, 2)=0.0628 (3, 1, 1)=0.0628 (4, 1)=0.0627\n", + " E 4: 4s loss=0.4210 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.093 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0628 (4, 1)=0.0627\n", + " E 5: 4s loss=0.4032 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.104 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0628 (4, 1)=0.0627\n", + " E 6: 4s loss=0.3938 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.091 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0628 (4, 1)=0.0627\n", + " E 7: 4s loss=0.3846 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.094 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0628 (4, 1)=0.0627\n", + " E 8: 4s loss=0.3765 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.094 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0628 (4, 1)=0.0627\n", + " E 9: 4s loss=0.3710 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.093 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0628 (4, 1)=0.0627\n", + " E10: 4s loss=0.3683 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.096 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0627 (4, 1)=0.0627\n", + " E11: 4s loss=0.3655 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.085 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0627 (4, 1)=0.0627\n", + " E12: 4s loss=0.3611 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.095 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0627 (4, 1)=0.0627\n", + " E13: 4s loss=0.3595 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.090 top: (3, 1, 1)=0.0629 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E14: 4s loss=0.3571 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E15: 4s loss=0.3549 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E16: 4s loss=0.3540 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.087 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E17: 4s loss=0.3527 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E18: 4s loss=0.3514 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.083 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E19: 4s loss=0.3508 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.078 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + " E20: 4s loss=0.3508 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (3, 1, 1)=0.0628 (1, 1, 1, 2)=0.0627 (4, 1)=0.0626\n", + "\n", + "=================================================================\n", + "FINAL PATH WEIGHT ANALYSIS (16 paths)\n", + "=================================================================\n", + " (3, 1, 1) 0.0628 █████ (3-way → indep → indep)\n", + " (1, 1, 1, 2) 0.0627 █████ (indep → indep → indep → pair)\n", + " (4, 1) 0.0626 █████ (4-way → indep)\n", + " (1, 4) 0.0626 █████ (indep → 4-way)\n", + " (2, 3) 0.0626 █████ (pair → 3-way)\n", + " (2, 2, 1) 0.0626 █████ (pair → pair → indep)\n", + " (2, 1, 1, 1) 0.0625 █████ (pair → indep → indep → indep)\n", + " (1, 1, 3) 0.0625 ████ (indep → indep → 3-way)\n", + " (1, 2, 1, 1) 0.0625 ████ (indep → pair → indep → indep)\n", + " (1, 1, 1, 1, 1) 0.0625 ████ (indep → indep → indep → indep → indep)\n", + " (2, 1, 2) 0.0624 ████ (pair → indep → pair)\n", + " (1, 3, 1) 0.0624 ████ (indep → 3-way → indep)\n", + " (3, 2) 0.0624 ████ (3-way → pair)\n", + " (1, 1, 2, 1) 0.0623 ████ (indep → indep → pair → indep)\n", + " (1, 2, 2) 0.0623 ████ (indep → pair → pair)\n", + " (5,) 0.0622 ████ (5-way)\n", + "\n", + " Weight spread: min=0.0622 max=0.0628 std=0.000154 range=0.000621\n", + " Uniform would be: 0.0625 = 0.0625\n", + "\n", + "=================================================================\n", + "CONSENSUS GEOMETRY\n", + "=================================================================\n", + " cos(output, bert ): 0.8415\n", + " cos(output, modern ): 0.8021\n", + " cos(output, roberta ): 0.8190\n", + " cos(output, albert ): 0.8024\n", + " cos(output, distil ): 0.8374\n", + " cos(output, centroid): 0.9085\n", + "\n", + " CV output: 0.0834\n", + " CV bert : 0.3988\n", + " CV modern : 0.3166\n", + " CV roberta : 0.3643\n", + " CV albert : 0.3431\n", + " CV distil : 0.3915\n", + "\n", + " Equidistance check (should be ~equal):\n", + " Range: 0.039444\n", + " Std: 0.016707\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n", + "\n", + "#################################################################\n", + "# SEED 12323 COMPLETE\n", + "#################################################################\n", + "\n", + "\n", + " Seed: 8675309\n", + "\n", + " Loading cached embeddings...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + "\n", + "=================================================================\n", + "PROCRUSTES ALIGNMENT (all → bert space)\n", + "=================================================================\n", + " bert : cos 1.0000 → 1.0000 (reference)\n", + " modern : cos -0.0025 → 0.4849\n", + " roberta : cos -0.0037 → 0.5138\n", + " albert : cos -0.0004 → 0.4888\n", + " distil : cos 0.8567 → 0.6557\n", + "\n", + " Pairwise cosines (post-alignment):\n", + " bert ↔ modern : 0.8357\n", + " bert ↔ roberta : 0.8685\n", + " bert ↔ albert : 0.8413\n", + " bert ↔ distil : 0.9314\n", + " modern ↔ roberta : 0.8040\n", + " modern ↔ albert : 0.7777\n", + " modern ↔ distil : 0.8224\n", + " roberta ↔ albert : 0.8039\n", + " roberta ↔ distil : 0.8528\n", + " albert ↔ distil : 0.8269\n", + "\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "MODEL: FiveAnchorModel + CompConv5d\n", + "=================================================================\n", + " Parameters: 16,910,352\n", + " Conv5d paths: 16 (2^4 = 16)\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " 5 InfoNCE + 5 SVD + 1 CV = 11 losses\n", + "=================================================================\n", + " E 1: 4s loss=3.5333 acc=0.926/0.910/0.918/0.916/0.923 val=1.000/1.000/1.000/0.998/1.000 cv=0.125 top: (4, 1)=0.0629 (1, 1, 2, 1)=0.0628 (2, 1, 2)=0.0627\n", + " E 2: 4s loss=0.5396 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.116 top: (4, 1)=0.0629 (1, 1, 2, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 3: 4s loss=0.4549 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.109 top: (4, 1)=0.0629 (1, 1, 2, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 4: 4s loss=0.4246 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.098 top: (1, 1, 2, 1)=0.0629 (4, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 5: 4s loss=0.4077 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.100 top: (1, 1, 2, 1)=0.0629 (4, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 6: 4s loss=0.3953 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.095 top: (1, 1, 2, 1)=0.0629 (4, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 7: 4s loss=0.3850 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.095 top: (1, 1, 2, 1)=0.0629 (4, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 8: 4s loss=0.3780 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.098 top: (1, 1, 2, 1)=0.0629 (4, 1)=0.0629 (2, 1, 2)=0.0627\n", + " E 9: 4s loss=0.3729 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.101 top: (4, 1)=0.0629 (1, 1, 2, 1)=0.0629 (2, 1, 2)=0.0626\n", + " E10: 4s loss=0.3697 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.090 top: (4, 1)=0.0629 (1, 1, 2, 1)=0.0628 (2, 3)=0.0626\n", + " E11: 4s loss=0.3650 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.082 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0626\n", + " E12: 4s loss=0.3632 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0626\n", + " E13: 4s loss=0.3610 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0626\n", + " E14: 4s loss=0.3587 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.073 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E15: 4s loss=0.3542 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.097 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E16: 4s loss=0.3537 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.086 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E17: 4s loss=0.3538 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.083 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E18: 4s loss=0.3514 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.084 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E19: 4s loss=0.3504 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.085 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + " E20: 4s loss=0.3506 acc=1.000/1.000/1.000/1.000/1.000 val=1.000/1.000/1.000/1.000/1.000 cv=0.077 top: (4, 1)=0.0628 (1, 1, 2, 1)=0.0628 (2, 3)=0.0627\n", + "\n", + "=================================================================\n", + "FINAL PATH WEIGHT ANALYSIS (16 paths)\n", + "=================================================================\n", + " (4, 1) 0.0628 █████ (4-way → indep)\n", + " (1, 1, 2, 1) 0.0628 █████ (indep → indep → pair → indep)\n", + " (2, 3) 0.0627 █████ (pair → 3-way)\n", + " (2, 1, 2) 0.0626 █████ (pair → indep → pair)\n", + " (3, 2) 0.0626 █████ (3-way → pair)\n", + " (1, 2, 1, 1) 0.0625 █████ (indep → pair → indep → indep)\n", + " (1, 3, 1) 0.0625 █████ (indep → 3-way → indep)\n", + " (1, 1, 1, 2) 0.0625 █████ (indep → indep → indep → pair)\n", + " (1, 4) 0.0625 ████ (indep → 4-way)\n", + " (1, 1, 3) 0.0625 ████ (indep → indep → 3-way)\n", + " (2, 2, 1) 0.0624 ████ (pair → pair → indep)\n", + " (2, 1, 1, 1) 0.0623 ████ (pair → indep → indep → indep)\n", + " (1, 1, 1, 1, 1) 0.0623 ████ (indep → indep → indep → indep → indep)\n", + " (3, 1, 1) 0.0623 ████ (3-way → indep → indep)\n", + " (5,) 0.0623 ████ (5-way)\n", + " (1, 2, 2) 0.0623 ████ (indep → pair → pair)\n", + "\n", + " Weight spread: min=0.0623 max=0.0628 std=0.000159 range=0.000530\n", + " Uniform would be: 0.0625 = 0.0625\n", + "\n", + "=================================================================\n", + "CONSENSUS GEOMETRY\n", + "=================================================================\n", + " cos(output, bert ): 0.8418\n", + " cos(output, modern ): 0.8024\n", + " cos(output, roberta ): 0.8190\n", + " cos(output, albert ): 0.8027\n", + " cos(output, distil ): 0.8374\n", + " cos(output, centroid): 0.9087\n", + "\n", + " CV output: 0.0835\n", + " CV bert : 0.4045\n", + " CV modern : 0.3322\n", + " CV roberta : 0.4071\n", + " CV albert : 0.3628\n", + " CV distil : 0.4071\n", + "\n", + " Equidistance check (should be ~equal):\n", + " Range: 0.039413\n", + " Std: 0.016641\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n", + "\n", + "#################################################################\n", + "# SEED 8675309 COMPLETE\n", + "#################################################################\n", + "\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# five bert distillation paradigm" + ], + "metadata": { + "id": "jnhe7llZaaGc" + } + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# DISTILLED CONSENSUS BERT: Standalone Caption Encoder\n", + "#\n", + "# Distills the five-BERT pentachoron consensus into a small standalone\n", + "# transformer that requires NO expert models at inference.\n", + "#\n", + "# Pipeline:\n", + "# 1. Generate consensus embeddings from cached five-BERT data\n", + "# 2. Train a small transformer to reproduce them from raw text\n", + "# 3. At inference: just the small model + tokenizer\n", + "#\n", + "# Target domain: image captions (CC12M), optimized for caption similarity\n", + "# ============================================================================\n", + "\n", + "import math\n", + "import os\n", + "import time\n", + "import json\n", + "from dataclasses import dataclass\n", + "\n", + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "from tqdm import tqdm\n", + "\n", + "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + "\n", + "print(\"=\" * 65)\n", + "print(\"DISTILLED CONSENSUS BERT\")\n", + "print(\"=\" * 65)\n", + "print(f\" Device: {DEVICE}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STEP 1: GENERATE CONSENSUS TARGETS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def load_consensus_targets():\n", + " \"\"\"\n", + " Load five-BERT embeddings, Procrustes align, run consensus model,\n", + " produce target embeddings for all captions.\n", + " \"\"\"\n", + " cache_dir = \"/home/claude/five_berts_cache\"\n", + " targets_path = os.path.join(cache_dir, \"consensus_targets.pt\")\n", + " captions_path = os.path.join(cache_dir, \"captions.json\")\n", + "\n", + " # Check if targets already generated\n", + " if os.path.exists(targets_path):\n", + " print(\"\\n Loading cached consensus targets...\")\n", + " data = torch.load(targets_path, weights_only=True)\n", + " with open(captions_path) as f:\n", + " captions = json.load(f)\n", + " print(f\" Targets: {data.shape}, Captions: {len(captions)}\")\n", + " return data, captions\n", + "\n", + " print(\"\\n Generating consensus targets from five-BERT cache...\")\n", + "\n", + " # Load raw embeddings\n", + " model_names = [\"bert\", \"modern\", \"roberta\", \"albert\", \"distil\"]\n", + " raw = {}\n", + " for name in model_names:\n", + " raw[name] = torch.load(os.path.join(cache_dir, f\"{name}.pt\"), weights_only=True)\n", + " print(f\" {name}: {raw[name].shape}\")\n", + "\n", + " with open(captions_path) as f:\n", + " captions = json.load(f)\n", + "\n", + " # Whitened Procrustes alignment (same as five_berts.py)\n", + " def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + " def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean\n", + " Tc = T - t_mean\n", + " N_s, D = Sc.shape\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " t_unwhiten = torch.linalg.pinv(t_whiten)\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten, \"target_unwhitener\": t_unwhiten,\n", + " }\n", + "\n", + " def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]\n", + " x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]\n", + " return x\n", + "\n", + " # Align all to bert space\n", + " ref = raw[\"bert\"]\n", + " aligned = {}\n", + " for name in model_names:\n", + " info = procrustes_align(raw[name], ref)\n", + " aligned[name] = apply_align(raw[name], info)\n", + " print(f\" Aligned {name}\")\n", + "\n", + " # Load consensus model\n", + " #from five_berts import FiveAnchorModel, CompConv5d, ConvPath, integer_compositions\n", + " model = FiveAnchorModel(768, 256, 5).to(DEVICE)\n", + " # Load best seed\n", + " for seed in [42, 43, 44]:\n", + " p = os.path.join(cache_dir, f\"five_model_seed{seed}.pt\")\n", + " if os.path.exists(p):\n", + " model.load_state_dict(torch.load(p, weights_only=True, map_location=DEVICE))\n", + " print(f\" Loaded consensus model (seed {seed})\")\n", + " break\n", + "\n", + " # Generate targets\n", + " model.eval()\n", + " all_targets = []\n", + " batch_size = 512\n", + " N = aligned[\"bert\"].shape[0]\n", + "\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, N, batch_size), desc=\" Generating targets\"):\n", + " j = min(i + batch_size, N)\n", + " anchors = [aligned[name][i:j].to(DEVICE) for name in model_names]\n", + " out = model(anchors)\n", + " all_targets.append(out.cpu())\n", + "\n", + " targets = torch.cat(all_targets)\n", + " print(f\" Consensus targets: {targets.shape}\")\n", + "\n", + " # Cache\n", + " torch.save(targets, targets_path)\n", + " print(f\" Saved to {targets_path}\")\n", + "\n", + " return targets, captions\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STEP 2: STUDENT MODEL\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class CaptionEncoder(nn.Module):\n", + " \"\"\"\n", + " Small standalone transformer for caption encoding.\n", + " No pretrained weights — trained from scratch on consensus targets.\n", + "\n", + " Architecture:\n", + " - Learned token embeddings (WordPiece vocab)\n", + " - Learned position embeddings (512 max)\n", + " - N transformer encoder layers\n", + " - Mean pooling → projection → L2-normalized output\n", + " \"\"\"\n", + " def __init__(self, vocab_size=30522, max_len=128, d_model=384,\n", + " n_heads=6, n_layers=4, d_ff=1536, output_dim=768,\n", + " dropout=0.1, pad_token_id=0):\n", + " super().__init__()\n", + " self.pad_token_id = pad_token_id\n", + " self.d_model = d_model\n", + "\n", + " self.token_emb = nn.Embedding(vocab_size, d_model, padding_idx=pad_token_id)\n", + " self.pos_emb = nn.Embedding(max_len, d_model)\n", + " self.emb_norm = nn.LayerNorm(d_model)\n", + " self.emb_drop = nn.Dropout(dropout)\n", + "\n", + " encoder_layer = nn.TransformerEncoderLayer(\n", + " d_model=d_model, nhead=n_heads, dim_feedforward=d_ff,\n", + " dropout=dropout, activation=\"gelu\", batch_first=True,\n", + " norm_first=True)\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n", + "\n", + " self.output_proj = nn.Sequential(\n", + " nn.Linear(d_model, d_model),\n", + " nn.GELU(),\n", + " nn.LayerNorm(d_model),\n", + " nn.Linear(d_model, output_dim),\n", + " )\n", + "\n", + " def forward(self, input_ids, attention_mask=None):\n", + " B, L = input_ids.shape\n", + " positions = torch.arange(L, device=input_ids.device).unsqueeze(0)\n", + "\n", + " x = self.token_emb(input_ids) + self.pos_emb(positions)\n", + " x = self.emb_drop(self.emb_norm(x))\n", + "\n", + " # Transformer mask\n", + " if attention_mask is not None:\n", + " src_key_padding_mask = ~attention_mask.bool()\n", + " else:\n", + " src_key_padding_mask = (input_ids == self.pad_token_id)\n", + "\n", + " x = self.encoder(x, src_key_padding_mask=src_key_padding_mask)\n", + "\n", + " # Mean pool over non-padding tokens\n", + " if attention_mask is not None:\n", + " mask = attention_mask.unsqueeze(-1).float()\n", + " else:\n", + " mask = (~src_key_padding_mask).unsqueeze(-1).float()\n", + " pooled = (x * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + "\n", + " return F.normalize(self.output_proj(pooled), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.084, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# TRAINING CONFIG\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "@dataclass\n", + "class TrainConfig:\n", + " # Student architecture\n", + " d_model: int = 384\n", + " n_heads: int = 6\n", + " n_layers: int = 4\n", + " d_ff: int = 1536\n", + " max_len: int = 128\n", + " output_dim: int = 768\n", + " dropout: float = 0.1\n", + "\n", + " # Training\n", + " epochs: int = 20\n", + " batch_size: int = 256\n", + " lr: float = 3e-4\n", + " weight_decay: float = 0.01\n", + " warmup_steps: int = 200\n", + " grad_clip: float = 1.0\n", + " seed: int = 42\n", + "\n", + " # Loss\n", + " nce_weight: float = 1.0\n", + " mse_weight: float = 1.0\n", + " cv_weight: float = 0.1\n", + " cv_target: float = 0.084 # consensus CV\n", + "\n", + " # Data\n", + " n_val: int = 2000\n", + "\n", + " # Paths\n", + " save_dir: str = \"/home/claude/distilled_consensus\"\n", + "\n", + "TCFG = TrainConfig()\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def train():\n", + " torch.manual_seed(TCFG.seed)\n", + " torch.cuda.manual_seed_all(TCFG.seed)\n", + " np.random.seed(TCFG.seed)\n", + "\n", + " # ── Targets ──\n", + " targets, captions = load_consensus_targets()\n", + "\n", + " # ── Tokenizer (BERT WordPiece — same vocab the consensus was built on) ──\n", + " from transformers import AutoTokenizer\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " print(f\"\\n Tokenizer: bert-base-uncased (vocab={tokenizer.vocab_size})\")\n", + "\n", + " # Pre-tokenize everything\n", + " print(\" Pre-tokenizing...\")\n", + " all_tokens = tokenizer(\n", + " captions, max_length=TCFG.max_len, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = all_tokens[\"input_ids\"]\n", + " attention_mask = all_tokens[\"attention_mask\"]\n", + "\n", + " # Token length stats\n", + " real_lens = attention_mask.sum(1).float()\n", + " print(f\" Token lengths: mean={real_lens.mean():.0f} \"\n", + " f\"median={real_lens.median():.0f} max={real_lens.max():.0f} \"\n", + " f\">{TCFG.max_len}: {(real_lens >= TCFG.max_len).float().mean():.1%}\")\n", + "\n", + " # Split\n", + " n_train = len(captions) - TCFG.n_val\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = targets[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = targets[n_train:].to(DEVICE)\n", + " print(f\" Train: {n_train}, Val: {TCFG.n_val}\")\n", + "\n", + " # ── Student model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"STUDENT MODEL\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student = CaptionEncoder(\n", + " vocab_size=tokenizer.vocab_size,\n", + " max_len=TCFG.max_len,\n", + " d_model=TCFG.d_model,\n", + " n_heads=TCFG.n_heads,\n", + " n_layers=TCFG.n_layers,\n", + " d_ff=TCFG.d_ff,\n", + " output_dim=TCFG.output_dim,\n", + " dropout=TCFG.dropout,\n", + " pad_token_id=tokenizer.pad_token_id,\n", + " ).to(DEVICE)\n", + "\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Architecture: {TCFG.n_layers} layers, {TCFG.d_model}-dim, \"\n", + " f\"{TCFG.n_heads} heads, {TCFG.d_ff} FFN\")\n", + " print(f\" Output: {TCFG.output_dim}-dim (matches consensus)\")\n", + " print(f\" Parameters: {n_params:,}\")\n", + "\n", + " # ── Optimizer ──\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=TCFG.lr,\n", + " weight_decay=TCFG.weight_decay)\n", + " n_batches = n_train // TCFG.batch_size\n", + " total_steps = n_batches * TCFG.epochs\n", + " scheduler = torch.optim.lr_scheduler.SequentialLR(\n", + " optimizer,\n", + " [torch.optim.lr_scheduler.LinearLR(optimizer, start_factor=0.01,\n", + " total_iters=TCFG.warmup_steps),\n", + " torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=max(total_steps - TCFG.warmup_steps, 1),\n", + " eta_min=1e-6)],\n", + " milestones=[TCFG.warmup_steps])\n", + "\n", + " os.makedirs(TCFG.save_dir, exist_ok=True)\n", + "\n", + " # ── Train ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({TCFG.epochs} epochs)\")\n", + " print(f\" Losses: InfoNCE + MSE + pentachoron CV (target={TCFG.cv_target})\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " all_metrics = {\"config\": vars(TCFG), \"epochs\": []}\n", + " best_val_cos = 0.0\n", + "\n", + " for epoch in range(TCFG.epochs):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " losses = {\"total\": 0, \"nce\": 0, \"mse\": 0, \"cv\": 0}\n", + " metrics = {\"acc\": 0, \"cos\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, TCFG.batch_size):\n", + " idx = perm[i:i+TCFG.batch_size]\n", + " if len(idx) < 8: continue\n", + "\n", + " ids = train_ids[idx]\n", + " mask = train_mask[idx]\n", + " tgt = train_targets[idx]\n", + "\n", + " emb = student(ids, mask)\n", + "\n", + " # InfoNCE: student should retrieve correct consensus target\n", + " l_nce, acc = infonce(emb, tgt)\n", + "\n", + " # MSE: direct regression on normalized embeddings\n", + " l_mse = F.mse_loss(emb, tgt)\n", + "\n", + " # CV: student embedding space should match consensus geometry\n", + " l_cv = cv_loss(emb, target=TCFG.cv_target)\n", + "\n", + " loss = (TCFG.nce_weight * l_nce +\n", + " TCFG.mse_weight * l_mse +\n", + " TCFG.cv_weight * l_cv)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), TCFG.grad_clip)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + "\n", + " losses[\"total\"] += loss.item()\n", + " losses[\"nce\"] += l_nce.item()\n", + " losses[\"mse\"] += l_mse.item()\n", + " metrics[\"acc\"] += acc\n", + " metrics[\"cos\"] += cos\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Validation\n", + " student.eval()\n", + " with torch.no_grad():\n", + " # Process val in chunks to avoid OOM\n", + " val_embs = []\n", + " for vi in range(0, TCFG.n_val, 512):\n", + " vj = min(vi + 512, TCFG.n_val)\n", + " ve = student(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(ve)\n", + " val_emb = torch.cat(val_embs)\n", + "\n", + " _, val_acc = infonce(val_emb, val_targets)\n", + " val_cos = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " val_cv = cv_metric(val_emb)\n", + "\n", + " summary = {\n", + " \"epoch\": epoch + 1, \"elapsed\": elapsed,\n", + " \"loss\": losses[\"total\"] / d,\n", + " \"train_acc\": metrics[\"acc\"] / d,\n", + " \"train_cos\": metrics[\"cos\"] / d,\n", + " \"val_acc\": val_acc,\n", + " \"val_cos\": val_cos,\n", + " \"val_cv\": val_cv,\n", + " }\n", + " all_metrics[\"epochs\"].append(summary)\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s \"\n", + " f\"loss={summary['loss']:.4f} \"\n", + " f\"t_acc={summary['train_acc']:.3f} t_cos={summary['train_cos']:.3f} \"\n", + " f\"v_acc={summary['val_acc']:.3f} v_cos={summary['val_cos']:.3f} \"\n", + " f\"v_cv={summary['val_cv']:.3f}\")\n", + "\n", + " # Save best\n", + " if val_cos > best_val_cos:\n", + " best_val_cos = val_cos\n", + " torch.save(student.state_dict(),\n", + " os.path.join(TCFG.save_dir, \"best_model.pt\"))\n", + "\n", + " # Save every 5 epochs\n", + " if (epoch + 1) % 5 == 0:\n", + " torch.save(student.state_dict(),\n", + " os.path.join(TCFG.save_dir, f\"model_e{epoch+1:02d}.pt\"))\n", + "\n", + " # ── Final save ──\n", + " torch.save(student.state_dict(),\n", + " os.path.join(TCFG.save_dir, \"final_model.pt\"))\n", + "\n", + " # Save tokenizer info for standalone usage\n", + " tokenizer.save_pretrained(os.path.join(TCFG.save_dir, \"tokenizer\"))\n", + "\n", + " with open(os.path.join(TCFG.save_dir, \"metrics.json\"), \"w\") as f:\n", + " json.dump(all_metrics, f, indent=2, default=str)\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # FINAL EVALUATION\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FINAL EVALUATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " # Load best\n", + " student.load_state_dict(\n", + " torch.load(os.path.join(TCFG.save_dir, \"best_model.pt\"),\n", + " weights_only=True, map_location=DEVICE))\n", + " student.eval()\n", + "\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, TCFG.n_val, 512):\n", + " vj = min(vi + 512, TCFG.n_val)\n", + " ve = student(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(ve)\n", + " val_emb = torch.cat(val_embs)\n", + "\n", + " # Retrieval metrics\n", + " sim = val_emb @ val_targets.T\n", + " labels = torch.arange(TCFG.n_val, device=DEVICE)\n", + " r1 = (sim.argmax(1) == labels).float().mean().item()\n", + " r5 = (sim.topk(5, dim=1).indices == labels.unsqueeze(1)).any(1).float().mean().item()\n", + " r10 = (sim.topk(10, dim=1).indices == labels.unsqueeze(1)).any(1).float().mean().item()\n", + "\n", + " # Cosine stats\n", + " cos_match = sim.diag().mean().item()\n", + " cos_random = (sim.sum() - sim.diag().sum()).item() / (TCFG.n_val**2 - TCFG.n_val)\n", + "\n", + " # CV\n", + " final_cv = cv_metric(val_emb)\n", + "\n", + " # Self-similarity (how well does student retrieve itself)\n", + " self_sim = val_emb @ val_emb.T\n", + " self_sim.fill_diagonal_(0)\n", + "\n", + " print(f\" Student → Consensus retrieval:\")\n", + " print(f\" R@1: {r1:.4f}\")\n", + " print(f\" R@5: {r5:.4f}\")\n", + " print(f\" R@10: {r10:.4f}\")\n", + " print(f\" Cosine similarity:\")\n", + " print(f\" Matched pairs: {cos_match:.4f}\")\n", + " print(f\" Random pairs: {cos_random:.4f}\")\n", + " print(f\" Geometry:\")\n", + " print(f\" CV: {final_cv:.4f} (target: {TCFG.cv_target})\")\n", + " print(f\" Model size:\")\n", + " print(f\" Parameters: {n_params:,}\")\n", + " size_mb = sum(p.numel() * p.element_size() for p in student.parameters()) / 1e6\n", + " print(f\" Size: {size_mb:.1f} MB\")\n", + "\n", + " # ── Standalone inference example ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"STANDALONE INFERENCE TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " test_captions = [\n", + " \"A cat sitting on a windowsill watching birds outside\",\n", + " \"A golden retriever playing fetch on the beach at sunset\",\n", + " \"A still life painting with flowers and fruit on a table\",\n", + " \"An aerial photograph of a city skyline at night\",\n", + " \"A child riding a bicycle through autumn leaves in a park\",\n", + " ]\n", + "\n", + " with torch.no_grad():\n", + " tokens = tokenizer(test_captions, max_length=TCFG.max_len,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " embeddings = student(tokens[\"input_ids\"], tokens[\"attention_mask\"])\n", + "\n", + " # Pairwise similarity\n", + " sim = embeddings @ embeddings.T\n", + " print(f\"\\n Pairwise cosine similarity:\")\n", + " for i in range(len(test_captions)):\n", + " for j in range(i+1, len(test_captions)):\n", + " print(f\" [{i}]↔[{j}]: {sim[i,j]:.3f} \"\n", + " f\"({test_captions[i][:40]}... ↔ {test_captions[j][:40]}...)\")\n", + "\n", + " print(f\"\\n Saved to: {TCFG.save_dir}/\")\n", + " print(f\" Files: best_model.pt, final_model.pt, tokenizer/, metrics.json\")\n", + " print(f\"\\n Standalone usage:\")\n", + " print(f\" model = CaptionEncoder(...)\")\n", + " print(f\" model.load_state_dict(torch.load('best_model.pt'))\")\n", + " print(f\" tokens = tokenizer(text, max_length=128, ...)\")\n", + " print(f\" embedding = model(tokens.input_ids, tokens.attention_mask)\")\n", + " print(f\" # → (B, 768) L2-normalized, consensus-aligned embedding\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " train()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "xQ_BobJFaZkc", + "outputId": "ca81f69b-e05f-47ba-80f2-c987bbd30fe6" + }, + "execution_count": 7, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "DISTILLED CONSENSUS BERT\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + " Generating consensus targets from five-BERT cache...\n", + " bert: torch.Size([20000, 768])\n", + " modern: torch.Size([20000, 768])\n", + " roberta: torch.Size([20000, 768])\n", + " albert: torch.Size([20000, 768])\n", + " distil: torch.Size([20000, 768])\n", + " Aligned bert\n", + " Aligned modern\n", + " Aligned roberta\n", + " Aligned albert\n", + " Aligned distil\n", + " Loaded consensus model (seed 42)\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + " Generating targets: 100%|██████████| 40/40 [00:00<00:00, 296.05it/s]\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " Consensus targets: torch.Size([20000, 768])\n", + " Saved to /home/claude/five_berts_cache/consensus_targets.pt\n", + "\n", + " Tokenizer: bert-base-uncased (vocab=30522)\n", + " Pre-tokenizing...\n", + " Token lengths: mean=92 median=100 max=128 >128: 22.5%\n", + " Train: 18000, Val: 2000\n", + "\n", + "=================================================================\n", + "STUDENT MODEL\n", + "=================================================================\n", + " Architecture: 4 layers, 384-dim, 6 heads, 1536 FFN\n", + " Output: 768-dim (matches consensus)\n", + " Parameters: 19,312,512\n", + "\n", + "=================================================================\n", + "TRAINING (20 epochs)\n", + " Losses: InfoNCE + MSE + pentachoron CV (target=0.084)\n", + "=================================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/tmp/ipykernel_1529/1100325579.py:178: UserWarning: enable_nested_tensor is True, but self.use_nested_tensor is False because encoder_layer.norm_first was True\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " E 1: 4s loss=4.4534 t_acc=0.163 t_cos=0.104 v_acc=0.236 v_cos=0.265 v_cv=0.230\n", + " E 2: 4s loss=2.0709 t_acc=0.715 t_cos=0.340 v_acc=0.642 v_cos=0.422 v_cv=0.196\n", + " E 3: 4s loss=1.1505 t_acc=0.908 t_cos=0.450 v_acc=0.833 v_cos=0.502 v_cv=0.169\n", + " E 4: 4s loss=0.7151 t_acc=0.967 t_cos=0.513 v_acc=0.913 v_cos=0.552 v_cv=0.146\n", + " E 5: 4s loss=0.4807 t_acc=0.986 t_cos=0.556 v_acc=0.932 v_cos=0.579 v_cv=0.133\n", + " E 6: 4s loss=0.3423 t_acc=0.994 t_cos=0.588 v_acc=0.962 v_cos=0.604 v_cv=0.130\n", + " E 7: 4s loss=0.2501 t_acc=0.997 t_cos=0.615 v_acc=0.970 v_cos=0.628 v_cv=0.127\n", + " E 8: 4s loss=0.1849 t_acc=0.999 t_cos=0.638 v_acc=0.981 v_cos=0.646 v_cv=0.113\n", + " E 9: 4s loss=0.1487 t_acc=1.000 t_cos=0.654 v_acc=0.985 v_cos=0.659 v_cv=0.109\n", + " E10: 4s loss=0.1271 t_acc=0.999 t_cos=0.665 v_acc=0.982 v_cos=0.665 v_cv=0.098\n", + " E11: 4s loss=0.1116 t_acc=1.000 t_cos=0.674 v_acc=0.987 v_cos=0.673 v_cv=0.103\n", + " E12: 4s loss=0.1027 t_acc=1.000 t_cos=0.681 v_acc=0.985 v_cos=0.675 v_cv=0.109\n", + " E13: 4s loss=0.0935 t_acc=1.000 t_cos=0.687 v_acc=0.988 v_cos=0.683 v_cv=0.105\n", + " E14: 4s loss=0.0880 t_acc=1.000 t_cos=0.691 v_acc=0.988 v_cos=0.684 v_cv=0.100\n", + " E15: 4s loss=0.0834 t_acc=1.000 t_cos=0.695 v_acc=0.989 v_cos=0.687 v_cv=0.112\n", + " E16: 4s loss=0.0804 t_acc=1.000 t_cos=0.698 v_acc=0.989 v_cos=0.688 v_cv=0.102\n", + " E17: 4s loss=0.0781 t_acc=1.000 t_cos=0.700 v_acc=0.989 v_cos=0.689 v_cv=0.099\n", + " E18: 4s loss=0.0767 t_acc=1.000 t_cos=0.701 v_acc=0.989 v_cos=0.690 v_cv=0.123\n", + " E19: 4s loss=0.0759 t_acc=1.000 t_cos=0.702 v_acc=0.989 v_cos=0.690 v_cv=0.103\n", + " E20: 4s loss=0.0760 t_acc=1.000 t_cos=0.702 v_acc=0.989 v_cos=0.690 v_cv=0.102\n", + "\n", + "=================================================================\n", + "FINAL EVALUATION\n", + "=================================================================\n", + " Student → Consensus retrieval:\n", + " R@1: 0.9890\n", + " R@5: 0.9985\n", + " R@10: 1.0000\n", + " Cosine similarity:\n", + " Matched pairs: 0.6905\n", + " Random pairs: 0.0005\n", + " Geometry:\n", + " CV: 0.1043 (target: 0.084)\n", + " Model size:\n", + " Parameters: 19,312,512\n", + " Size: 77.3 MB\n", + "\n", + "=================================================================\n", + "STANDALONE INFERENCE TEST\n", + "=================================================================\n", + "\n", + " Pairwise cosine similarity:\n", + " [0]↔[1]: 0.677 (A cat sitting on a windowsill watching b... ↔ A golden retriever playing fetch on the ...)\n", + " [0]↔[2]: 0.404 (A cat sitting on a windowsill watching b... ↔ A still life painting with flowers and f...)\n", + " [0]↔[3]: 0.421 (A cat sitting on a windowsill watching b... ↔ An aerial photograph of a city skyline a...)\n", + " [0]↔[4]: 0.595 (A cat sitting on a windowsill watching b... ↔ A child riding a bicycle through autumn ...)\n", + " [1]↔[2]: 0.354 (A golden retriever playing fetch on the ... ↔ A still life painting with flowers and f...)\n", + " [1]↔[3]: 0.353 (A golden retriever playing fetch on the ... ↔ An aerial photograph of a city skyline a...)\n", + " [1]↔[4]: 0.508 (A golden retriever playing fetch on the ... ↔ A child riding a bicycle through autumn ...)\n", + " [2]↔[3]: 0.363 (A still life painting with flowers and f... ↔ An aerial photograph of a city skyline a...)\n", + " [2]↔[4]: 0.357 (A still life painting with flowers and f... ↔ A child riding a bicycle through autumn ...)\n", + " [3]↔[4]: 0.353 (An aerial photograph of a city skyline a... ↔ A child riding a bicycle through autumn ...)\n", + "\n", + " Saved to: /home/claude/distilled_consensus/\n", + " Files: best_model.pt, final_model.pt, tokenizer/, metrics.json\n", + "\n", + " Standalone usage:\n", + " model = CaptionEncoder(...)\n", + " model.load_state_dict(torch.load('best_model.pt'))\n", + " tokens = tokenizer(text, max_length=128, ...)\n", + " embedding = model(tokens.input_ids, tokens.attention_mask)\n", + " # → (B, 768) L2-normalized, consensus-aligned embedding\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# distill 2" + ], + "metadata": { + "id": "GSm-HEucekY8" + } + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# DISTILLED CONSENSUS BERT — 200K Scale\n", + "#\n", + "# Self-contained pipeline:\n", + "# 1. Extract 5 BERT-family embeddings on 200K CC12M captions\n", + "# 2. Whitened Procrustes alignment\n", + "# 3. Generate consensus targets (centroid of aligned embeddings)\n", + "# 4. Train small standalone transformer from scratch\n", + "# 5. No expert models needed at inference\n", + "# ============================================================================\n", + "\n", + "import math\n", + "import os\n", + "import time\n", + "import json\n", + "from dataclasses import dataclass\n", + "\n", + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "from tqdm import tqdm\n", + "\n", + "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + "\n", + "MODELS = [\n", + " (\"google-bert/bert-base-uncased\", \"bert\"),\n", + " (\"answerdotai/ModernBERT-base\", \"modern\"),\n", + " (\"FacebookAI/roberta-base\", \"roberta\"),\n", + " (\"albert/albert-base-v2\", \"albert\"),\n", + " (\"distilbert/distilbert-base-uncased\", \"distil\"),\n", + "]\n", + "\n", + "@dataclass\n", + "class Config:\n", + " # Data\n", + " n_samples: int = 200000\n", + " n_val: int = 5000\n", + " min_caption_len: int = 50\n", + " extract_batch: int = 128\n", + " extract_max_len: int = 512\n", + " cache_dir: str = \"/home/claude/consensus_200k\"\n", + "\n", + " # Student architecture\n", + " d_model: int = 384\n", + " n_heads: int = 6\n", + " n_layers: int = 6\n", + " d_ff: int = 1536\n", + " max_len: int = 128\n", + " output_dim: int = 768\n", + " dropout: float = 0.1\n", + "\n", + " # Training\n", + " epochs: int = 30\n", + " batch_size: int = 256\n", + " lr: float = 3e-4\n", + " weight_decay: float = 0.01\n", + " warmup_steps: int = 500\n", + " grad_clip: float = 1.0\n", + " seed: int = 42\n", + "\n", + " # Loss\n", + " nce_weight: float = 1.0\n", + " mse_weight: float = 1.0\n", + " cv_weight: float = 0.1\n", + " cv_target: float = 0.084\n", + "\n", + "CFG = Config()\n", + "\n", + "print(\"=\" * 65)\n", + "print(\"DISTILLED CONSENSUS BERT — 200K Scale\")\n", + "print(\"=\" * 65)\n", + "print(f\" Device: {DEVICE}\")\n", + "print(f\" Samples: {CFG.n_samples:,}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def load_captions(n, min_len=50):\n", + " from datasets import load_dataset\n", + " print(f\"\\n Loading captions (n={n:,})...\")\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > min_len:\n", + " captions.append(cap)\n", + " if len(captions) >= n:\n", + " break\n", + " print(f\" Got {len(captions):,} captions\")\n", + " return captions\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def extract_one(model_name, short_name, captions, max_len, batch_size):\n", + " from transformers import AutoModel, AutoTokenizer\n", + " print(f\"\\n Extracting: {short_name} ({model_name})...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " dim = model.config.hidden_size\n", + " n_params = sum(p.numel() for p in model.parameters())\n", + " print(f\" dim={dim}, {n_params:,} params\")\n", + "\n", + " all_emb = []\n", + " for i in tqdm(range(0, len(captions), batch_size), desc=f\" {short_name}\"):\n", + " batch = captions[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + "\n", + " emb = torch.cat(all_emb)\n", + " print(f\" Shape: {emb.shape}\")\n", + " del model\n", + " torch.cuda.empty_cache()\n", + " return emb\n", + "\n", + "\n", + "def extract_all():\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " caps_path = os.path.join(CFG.cache_dir, \"captions.json\")\n", + "\n", + " all_cached = all(\n", + " os.path.exists(os.path.join(CFG.cache_dir, f\"{s}.pt\"))\n", + " for _, s in MODELS)\n", + "\n", + " if all_cached and os.path.exists(caps_path):\n", + " print(\"\\n Loading cached embeddings...\")\n", + " embeds = {}\n", + " for _, short in MODELS:\n", + " embeds[short] = torch.load(\n", + " os.path.join(CFG.cache_dir, f\"{short}.pt\"), weights_only=True)\n", + " print(f\" {short}: {embeds[short].shape}\")\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " return embeds, captions\n", + "\n", + " captions = load_captions(CFG.n_samples, CFG.min_caption_len)\n", + "\n", + " embeds = {}\n", + " for model_name, short in MODELS:\n", + " emb = extract_one(model_name, short, captions,\n", + " CFG.extract_max_len, CFG.extract_batch)\n", + " if emb.shape[1] != 768:\n", + " if emb.shape[1] < 768:\n", + " emb = F.pad(emb, (0, 768 - emb.shape[1]))\n", + " else:\n", + " emb = emb[:, :768]\n", + " embeds[short] = emb\n", + " torch.save(emb, os.path.join(CFG.cache_dir, f\"{short}.pt\"))\n", + "\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " return embeds, captions\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# WHITENED PROCRUSTES + CONSENSUS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "\n", + "def procrustes_align(source, target, n_align=10000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean\n", + " Tc = T - t_mean\n", + " N_s = Sc.shape[0]\n", + "\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + "\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + "\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + "\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + "\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + "\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]\n", + " x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]\n", + " return x\n", + "\n", + "\n", + "def generate_consensus(embeds):\n", + " \"\"\"Align all to bert space, take normalized centroid as target.\"\"\"\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"WHITENED PROCRUSTES ALIGNMENT + CONSENSUS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref_name = \"bert\"\n", + " names = [s for _, s in MODELS]\n", + " aligned = {}\n", + "\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], embeds[ref_name])\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " label = \" (ref)\" if name == ref_name else \"\"\n", + " print(f\" {name:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}{label}\")\n", + "\n", + " # Consensus = normalized centroid of all 5 aligned embeddings\n", + " # This is what the five-BERT experiment proved: the centroid IS the consensus\n", + " # to three decimal places regardless of seed. No learned model needed.\n", + " centroid = sum(aligned[n] for n in names) / len(names)\n", + " consensus = F.normalize(centroid, dim=-1)\n", + "\n", + " # Verify geometry\n", + " N_check = min(5000, consensus.shape[0])\n", + " for name in names:\n", + " cos = F.cosine_similarity(\n", + " consensus[:N_check], aligned[name][:N_check], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name:10s}): {cos:.4f}\")\n", + "\n", + " return consensus\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STUDENT MODEL\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class CaptionEncoder(nn.Module):\n", + " def __init__(self, vocab_size=30522, max_len=128, d_model=384,\n", + " n_heads=6, n_layers=6, d_ff=1536, output_dim=768,\n", + " dropout=0.1, pad_token_id=0):\n", + " super().__init__()\n", + " self.pad_token_id = pad_token_id\n", + " self.token_emb = nn.Embedding(vocab_size, d_model, padding_idx=pad_token_id)\n", + " self.pos_emb = nn.Embedding(max_len, d_model)\n", + " self.emb_norm = nn.LayerNorm(d_model)\n", + " self.emb_drop = nn.Dropout(dropout)\n", + "\n", + " encoder_layer = nn.TransformerEncoderLayer(\n", + " d_model=d_model, nhead=n_heads, dim_feedforward=d_ff,\n", + " dropout=dropout, activation=\"gelu\", batch_first=True,\n", + " norm_first=True)\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n", + "\n", + " self.output_proj = nn.Sequential(\n", + " nn.Linear(d_model, d_model),\n", + " nn.GELU(),\n", + " nn.LayerNorm(d_model),\n", + " nn.Linear(d_model, output_dim),\n", + " )\n", + "\n", + " def forward(self, input_ids, attention_mask=None):\n", + " B, L = input_ids.shape\n", + " positions = torch.arange(L, device=input_ids.device).unsqueeze(0)\n", + " x = self.token_emb(input_ids) + self.pos_emb(positions)\n", + " x = self.emb_drop(self.emb_norm(x))\n", + "\n", + " if attention_mask is not None:\n", + " kpm = ~attention_mask.bool()\n", + " else:\n", + " kpm = (input_ids == self.pad_token_id)\n", + "\n", + " x = self.encoder(x, src_key_padding_mask=kpm)\n", + "\n", + " if attention_mask is not None:\n", + " mask = attention_mask.unsqueeze(-1).float()\n", + " else:\n", + " mask = (~kpm).unsqueeze(-1).float()\n", + " pooled = (x * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + "\n", + " return F.normalize(self.output_proj(pooled), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.084, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def train():\n", + " torch.manual_seed(CFG.seed)\n", + " torch.cuda.manual_seed_all(CFG.seed)\n", + " np.random.seed(CFG.seed)\n", + "\n", + " # ── Extract + Align + Consensus ──\n", + " embeds, captions = extract_all()\n", + " consensus = generate_consensus(embeds)\n", + "\n", + " # Free the raw embeddings\n", + " del embeds\n", + " torch.cuda.empty_cache()\n", + " import gc; gc.collect()\n", + "\n", + " # ── Tokenize ──\n", + " from transformers import AutoTokenizer\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " print(f\"\\n Tokenizer: bert-base-uncased (vocab={tokenizer.vocab_size})\")\n", + "\n", + " print(\" Pre-tokenizing...\")\n", + " # Tokenize in chunks to avoid memory issues\n", + " all_ids, all_masks = [], []\n", + " chunk = 50000\n", + " for i in tqdm(range(0, len(captions), chunk), desc=\" Tokenizing\"):\n", + " j = min(i + chunk, len(captions))\n", + " tokens = tokenizer(captions[i:j], max_length=CFG.max_len,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\")\n", + " all_ids.append(tokens[\"input_ids\"])\n", + " all_masks.append(tokens[\"attention_mask\"])\n", + "\n", + " input_ids = torch.cat(all_ids)\n", + " attention_mask = torch.cat(all_masks)\n", + "\n", + " real_lens = attention_mask.sum(1).float()\n", + " print(f\" Token lengths: mean={real_lens.mean():.0f} \"\n", + " f\"median={real_lens.median():.0f} \"\n", + " f\">{CFG.max_len}: {(real_lens >= CFG.max_len).float().mean():.1%}\")\n", + "\n", + " # Split\n", + " n_train = len(captions) - CFG.n_val\n", + " print(f\" Train: {n_train:,}, Val: {CFG.n_val:,}\")\n", + "\n", + " # Move to GPU\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " # ── Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"STUDENT MODEL\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student = CaptionEncoder(\n", + " vocab_size=tokenizer.vocab_size,\n", + " max_len=CFG.max_len,\n", + " d_model=CFG.d_model,\n", + " n_heads=CFG.n_heads,\n", + " n_layers=CFG.n_layers,\n", + " d_ff=CFG.d_ff,\n", + " output_dim=CFG.output_dim,\n", + " dropout=CFG.dropout,\n", + " pad_token_id=tokenizer.pad_token_id,\n", + " ).to(DEVICE)\n", + "\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Architecture: {CFG.n_layers}L, {CFG.d_model}d, {CFG.n_heads}h, {CFG.d_ff} FFN\")\n", + " print(f\" Output: {CFG.output_dim}-dim (consensus space)\")\n", + " print(f\" Parameters: {n_params:,}\")\n", + " size_mb = sum(p.numel() * p.element_size() for p in student.parameters()) / 1e6\n", + " print(f\" Size: {size_mb:.1f} MB\")\n", + "\n", + " # ── Optimizer ──\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=CFG.lr,\n", + " weight_decay=CFG.weight_decay)\n", + " n_batches = n_train // CFG.batch_size\n", + " total_steps = n_batches * CFG.epochs\n", + " scheduler = torch.optim.lr_scheduler.SequentialLR(\n", + " optimizer,\n", + " [torch.optim.lr_scheduler.LinearLR(optimizer, start_factor=0.01,\n", + " total_iters=CFG.warmup_steps),\n", + " torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=max(total_steps - CFG.warmup_steps, 1),\n", + " eta_min=1e-6)],\n", + " milestones=[CFG.warmup_steps])\n", + "\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " save_dir = os.path.join(CFG.cache_dir, \"student\")\n", + " os.makedirs(save_dir, exist_ok=True)\n", + "\n", + " # ── Train ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({CFG.epochs} epochs, {n_batches} batches/epoch)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " all_metrics = {\"config\": {k: str(v) for k, v in vars(CFG).items()}, \"epochs\": []}\n", + " best_val_cos = 0.0\n", + "\n", + " for epoch in range(CFG.epochs):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " losses = {\"total\": 0, \"nce\": 0, \"mse\": 0}\n", + " metrics = {\"acc\": 0, \"cos\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, CFG.batch_size):\n", + " idx = perm[i:i+CFG.batch_size]\n", + " if len(idx) < 8: continue\n", + "\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + "\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=CFG.cv_target)\n", + "\n", + " loss = CFG.nce_weight * l_nce + CFG.mse_weight * l_mse + CFG.cv_weight * l_cv\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), CFG.grad_clip)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + "\n", + " losses[\"total\"] += loss.item()\n", + " losses[\"nce\"] += l_nce.item()\n", + " losses[\"mse\"] += l_mse.item()\n", + " metrics[\"acc\"] += acc\n", + " metrics[\"cos\"] += cos\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Val\n", + " student.eval()\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, CFG.n_val, 512):\n", + " vj = min(vi + 512, CFG.n_val)\n", + " ve = student(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(ve)\n", + " val_emb = torch.cat(val_embs)\n", + " _, val_acc = infonce(val_emb[:2000], val_targets[:2000])\n", + " val_cos = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " val_cv = cv_metric(val_emb[:2000])\n", + "\n", + " summary = {\n", + " \"epoch\": epoch + 1, \"elapsed\": elapsed,\n", + " \"loss\": losses[\"total\"] / d,\n", + " \"train_acc\": metrics[\"acc\"] / d,\n", + " \"train_cos\": metrics[\"cos\"] / d,\n", + " \"val_acc\": val_acc, \"val_cos\": val_cos, \"val_cv\": val_cv,\n", + " }\n", + " all_metrics[\"epochs\"].append(summary)\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s \"\n", + " f\"loss={summary['loss']:.4f} \"\n", + " f\"t_acc={summary['train_acc']:.3f} t_cos={summary['train_cos']:.3f} \"\n", + " f\"v_acc={summary['val_acc']:.3f} v_cos={summary['val_cos']:.3f} \"\n", + " f\"v_cv={summary['val_cv']:.3f}\")\n", + "\n", + " if val_cos > best_val_cos:\n", + " best_val_cos = val_cos\n", + " torch.save(student.state_dict(), os.path.join(save_dir, \"best_model.pt\"))\n", + "\n", + " if (epoch + 1) % 10 == 0:\n", + " torch.save(student.state_dict(),\n", + " os.path.join(save_dir, f\"model_e{epoch+1:02d}.pt\"))\n", + "\n", + " # Final save\n", + " torch.save(student.state_dict(), os.path.join(save_dir, \"final_model.pt\"))\n", + " tokenizer.save_pretrained(os.path.join(save_dir, \"tokenizer\"))\n", + " with open(os.path.join(save_dir, \"metrics.json\"), \"w\") as f:\n", + " json.dump(all_metrics, f, indent=2, default=str)\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # FINAL EVAL\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FINAL EVALUATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.load_state_dict(\n", + " torch.load(os.path.join(save_dir, \"best_model.pt\"),\n", + " weights_only=True, map_location=DEVICE))\n", + " student.eval()\n", + "\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, CFG.n_val, 512):\n", + " vj = min(vi + 512, CFG.n_val)\n", + " ve = student(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(ve)\n", + " val_emb = torch.cat(val_embs)\n", + "\n", + " # Retrieval (on 2K subset for memory)\n", + " sub = min(2000, CFG.n_val)\n", + " sim = val_emb[:sub] @ val_targets[:sub].T\n", + " labels = torch.arange(sub, device=DEVICE)\n", + " r1 = (sim.argmax(1) == labels).float().mean().item()\n", + " r5 = (sim.topk(5, dim=1).indices == labels.unsqueeze(1)).any(1).float().mean().item()\n", + " r10 = (sim.topk(10, dim=1).indices == labels.unsqueeze(1)).any(1).float().mean().item()\n", + "\n", + " cos_match = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " final_cv = cv_metric(val_emb[:2000])\n", + "\n", + " print(f\" Retrieval (student → consensus):\")\n", + " print(f\" R@1: {r1:.4f}\")\n", + " print(f\" R@5: {r5:.4f}\")\n", + " print(f\" R@10: {r10:.4f}\")\n", + " print(f\" Cosine: {cos_match:.4f}\")\n", + " print(f\" CV: {final_cv:.4f} (target: {CFG.cv_target})\")\n", + " print(f\" Model: {n_params:,} params, {size_mb:.1f} MB\")\n", + "\n", + " # Standalone test\n", + " print(f\"\\n Standalone similarity test:\")\n", + " test = [\n", + " \"A cat sitting on a windowsill watching birds\",\n", + " \"A golden retriever playing fetch on the beach\",\n", + " \"A still life painting with flowers and fruit\",\n", + " \"An aerial photograph of a city skyline at night\",\n", + " \"A child riding a bicycle through autumn leaves\",\n", + " ]\n", + " with torch.no_grad():\n", + " tok = tokenizer(test, max_length=CFG.max_len, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " embs = student(tok[\"input_ids\"], tok[\"attention_mask\"])\n", + " sim = embs @ embs.T\n", + " for i in range(len(test)):\n", + " for j in range(i+1, len(test)):\n", + " print(f\" [{i}]↔[{j}]: {sim[i,j]:.3f} \"\n", + " f\"({test[i][:35]}↔{test[j][:35]})\")\n", + "\n", + " print(f\"\\n Saved to: {save_dir}/\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " train()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "0eaf1b9b3faf4564b2ea751d3292a62b", + "61f98e3de9f54cd8bb6ac5abf94c4547", + "e81f9655321c435f898aaf1d2bef667c", + "ff6d3c3b5f6c45a58746baf241ab8f69", + "6311dfcc5ede419cbebd064cb42af75a", + "5f3e8530131045459d831a4cd75d752d", + "80acec4b777947bf9481d1b6292178c0", + "1a6e19bfe5244157a110d80615352827", + "96d62c3895ec40449c75d8527dc6f1e5", + "3e71024a339a4938a94205fcbb4ba530", + "2038b2fbe7d846178ecf6cf083ca55de", + "abdd244173644b379041034854e908fa", + "122a0e0816f945ceb1942cd3d10f51f6", + "6ef64c471dc94f609440fb4c4ed0f9e4", + "449a39840fac45a9bf33b016a4ecde66", + "af961b31ee0e4cabac70dd4e61cd2f05", + "e372742ba260499b875618da8b0b62df", + "a3d8b3f6224a4f5e85a1b1a8fc116cc0", + "3273f54a6fcc4e82aee577b2b508b493", + "631ea49dbaa3495688f16e0b27c90610", + "beed45db486d40dbb70fd5a6713fcd5d", + "4a2611a199da4fef9dafd0c3934e02fe", + "defad155b30b4076ae5431e8b299a148", + "cac0d30d1ba848f28e2722a7baddf6fe", + "38b121d084cc40e58727c89299acfef3", + "bbe4c87a57804f3d9332abc541c938ed", + "7754907597864c7386d8234500b1df3b", + "34c8e8c9f27f4384b33a50f37f0dd97c", + "65d3b06a14c74e9f944194dbe682d070", + "8ff7e87e23e048ce90ca973a803874e9", + "07d12ebd42d1426091b26eeb77a17d97", + "d77868ac7580404490ca0f80ec607cc3", + "d666624772a6477d90dce56414e97ac9", + "8eb0acb8d0174d89ad3869e17c73468c", + "ba39be331d814cb4ac763e3f9fde1032", + "132de8e2ddd541269a810f44cc6fb971", + "e4ce3e29132a45649817df4f7e8341e2", + "3bf5704b4d6942909ad163b3b91318af", + "87dbb92120d84e418a9ba3770b6ddb73", + "e4eda51e8e614607901352ff01958332", + "e0234180e8644ba99c472de64827ce69", + "99482875446a40d488ded2b1522820f9", + "a177a905339d494a9938dee8e6446983", + "c5ed7bba1e6a4cf18e5bc3bda01bcada", + "380a95ce1d17446fb7a8370f95e2d62d", + "17ba56553c1545dc8899a7ff580e3b4a", + "43e22d4a71a74918be6c8c4a6dccd928", + "96ddc7a8e7f4482aa863aae9194bc2fa", + "698be3985bc14d93bdb4f2d8ad7114f9", + "07154f164e574d4ea174183bab680de6", + "b943e03d5d094af2abf688177557d1b3", + "5daca214c00c40a08aaa0ccb44406d93", + "d473e9f0063341a4871d5c507d012fe7", + "4a39b4a3ddb5468ebeb0f595732d681f", + "92a8707d70614dada558e2cc8c90b1d7" + ] + }, + "id": "lVZLPcTlel2K", + "outputId": "d4a6f306-5d25-4173-a938-bccb6607cf18" + }, + "execution_count": 8, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "DISTILLED CONSENSUS BERT — 200K Scale\n", + "=================================================================\n", + " Device: cuda\n", + " Samples: 200,000\n", + "\n", + " Loading captions (n=200,000)...\n", + " Got 200,000 captions\n", + "\n", + " Extracting: bert (google-bert/bert-base-uncased)...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00128: 22.8%\n", + " Train: 195,000, Val: 5,000\n", + "\n", + "=================================================================\n", + "STUDENT MODEL\n", + "=================================================================\n", + " Architecture: 6L, 384d, 6h, 1536 FFN\n", + " Output: 768-dim (consensus space)\n", + " Parameters: 22,861,440\n", + " Size: 91.4 MB\n", + "\n", + "=================================================================\n", + "TRAINING (30 epochs, 761 batches/epoch)\n", + "=================================================================\n", + " E 1: 61s loss=1.3769 t_acc=0.787 t_cos=0.517 v_acc=0.979 v_cos=0.702 v_cv=0.141\n", + " E 2: 61s loss=0.1657 t_acc=0.998 t_cos=0.722 v_acc=0.994 v_cos=0.745 v_cv=0.112\n", + " E 3: 61s loss=0.0995 t_acc=1.000 t_cos=0.752 v_acc=0.997 v_cos=0.766 v_cv=0.108\n", + " E 4: 61s loss=0.0778 t_acc=1.000 t_cos=0.766 v_acc=0.997 v_cos=0.771 v_cv=0.106\n", + " E 5: 61s loss=0.0658 t_acc=1.000 t_cos=0.776 v_acc=0.999 v_cos=0.776 v_cv=0.103\n", + " E 6: 61s loss=0.0580 t_acc=1.000 t_cos=0.782 v_acc=0.998 v_cos=0.789 v_cv=0.113\n", + " E 7: 61s loss=0.0523 t_acc=1.000 t_cos=0.789 v_acc=0.999 v_cos=0.793 v_cv=0.088\n", + " E 8: 61s loss=0.0479 t_acc=1.000 t_cos=0.793 v_acc=1.000 v_cos=0.795 v_cv=0.101\n", + " E 9: 61s loss=0.0449 t_acc=1.000 t_cos=0.797 v_acc=1.000 v_cos=0.798 v_cv=0.090\n", + " E10: 61s loss=0.0419 t_acc=1.000 t_cos=0.801 v_acc=1.000 v_cos=0.801 v_cv=0.084\n", + " E11: 61s loss=0.0399 t_acc=1.000 t_cos=0.805 v_acc=1.000 v_cos=0.807 v_cv=0.094\n", + " E12: 61s loss=0.0379 t_acc=1.000 t_cos=0.809 v_acc=1.000 v_cos=0.810 v_cv=0.099\n", + " E13: 61s loss=0.0360 t_acc=1.000 t_cos=0.812 v_acc=1.000 v_cos=0.813 v_cv=0.099\n", + " E14: 61s loss=0.0345 t_acc=1.000 t_cos=0.814 v_acc=1.000 v_cos=0.814 v_cv=0.092\n", + " E15: 61s loss=0.0332 t_acc=1.000 t_cos=0.817 v_acc=1.000 v_cos=0.815 v_cv=0.088\n", + " E16: 61s loss=0.0321 t_acc=1.000 t_cos=0.819 v_acc=1.000 v_cos=0.817 v_cv=0.102\n", + " E17: 61s loss=0.0310 t_acc=1.000 t_cos=0.821 v_acc=1.000 v_cos=0.820 v_cv=0.092\n", + " E18: 61s loss=0.0300 t_acc=1.000 t_cos=0.824 v_acc=1.000 v_cos=0.822 v_cv=0.091\n", + " E19: 61s loss=0.0292 t_acc=1.000 t_cos=0.827 v_acc=1.000 v_cos=0.822 v_cv=0.086\n", + " E20: 61s loss=0.0284 t_acc=1.000 t_cos=0.828 v_acc=1.000 v_cos=0.827 v_cv=0.091\n", + " E21: 61s loss=0.0277 t_acc=1.000 t_cos=0.830 v_acc=1.000 v_cos=0.826 v_cv=0.087\n", + " E22: 61s loss=0.0272 t_acc=1.000 t_cos=0.831 v_acc=1.000 v_cos=0.829 v_cv=0.093\n", + " E23: 61s loss=0.0264 t_acc=1.000 t_cos=0.833 v_acc=1.000 v_cos=0.830 v_cv=0.081\n", + " E24: 61s loss=0.0259 t_acc=1.000 t_cos=0.836 v_acc=1.000 v_cos=0.829 v_cv=0.090\n", + " E25: 61s loss=0.0257 t_acc=1.000 t_cos=0.836 v_acc=1.000 v_cos=0.830 v_cv=0.082\n", + " E26: 61s loss=0.0251 t_acc=1.000 t_cos=0.838 v_acc=1.000 v_cos=0.831 v_cv=0.086\n", + " E27: 61s loss=0.0251 t_acc=1.000 t_cos=0.838 v_acc=1.000 v_cos=0.832 v_cv=0.090\n", + " E28: 61s loss=0.0248 t_acc=1.000 t_cos=0.839 v_acc=1.000 v_cos=0.833 v_cv=0.091\n", + " E29: 61s loss=0.0247 t_acc=1.000 t_cos=0.840 v_acc=1.000 v_cos=0.833 v_cv=0.085\n", + " E30: 61s loss=0.0249 t_acc=1.000 t_cos=0.840 v_acc=1.000 v_cos=0.833 v_cv=0.090\n", + "\n", + "=================================================================\n", + "FINAL EVALUATION\n", + "=================================================================\n", + " Retrieval (student → consensus):\n", + " R@1: 1.0000\n", + " R@5: 1.0000\n", + " R@10: 1.0000\n", + " Cosine: 0.8333\n", + " CV: 0.0894 (target: 0.084)\n", + " Model: 22,861,440 params, 91.4 MB\n", + "\n", + " Standalone similarity test:\n", + " [0]↔[1]: 0.702 (A cat sitting on a windowsill watch↔A golden retriever playing fetch on)\n", + " [0]↔[2]: 0.356 (A cat sitting on a windowsill watch↔A still life painting with flowers )\n", + " [0]↔[3]: 0.421 (A cat sitting on a windowsill watch↔An aerial photograph of a city skyl)\n", + " [0]↔[4]: 0.445 (A cat sitting on a windowsill watch↔A child riding a bicycle through au)\n", + " [1]↔[2]: 0.340 (A golden retriever playing fetch on↔A still life painting with flowers )\n", + " [1]↔[3]: 0.407 (A golden retriever playing fetch on↔An aerial photograph of a city skyl)\n", + " [1]↔[4]: 0.396 (A golden retriever playing fetch on↔A child riding a bicycle through au)\n", + " [2]↔[3]: 0.381 (A still life painting with flowers ↔An aerial photograph of a city skyl)\n", + " [2]↔[4]: 0.471 (A still life painting with flowers ↔A child riding a bicycle through au)\n", + " [3]↔[4]: 0.382 (An aerial photograph of a city skyl↔A child riding a bicycle through au)\n", + "\n", + " Saved to: /home/claude/consensus_200k/student/\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# BENCHMARK: Distilled Consensus Student vs Individual BERTs\n", + "#\n", + "# Tests:\n", + "# 1. STS-B (Semantic Textual Similarity Benchmark) — Spearman correlation\n", + "# 2. SICK-R (Sentences Involving Compositional Knowledge) — Spearman\n", + "# 3. Retrieval precision on held-out consensus targets\n", + "#\n", + "# Compares:\n", + "# - Distilled student (19-23M params, no pretrained weights)\n", + "# - BERT-base-uncased (110M params)\n", + "# - ModernBERT-base (149M params)\n", + "# - RoBERTa-base (125M params)\n", + "# - ALBERT-base-v2 (12M params)\n", + "# - DistilBERT-base (66M params)\n", + "#\n", + "# All models evaluated on mean-pooled embeddings → cosine similarity\n", + "# ============================================================================\n", + "\n", + "import os\n", + "import json\n", + "import torch\n", + "import torch.nn.functional as F\n", + "import numpy as np\n", + "from scipy.stats import spearmanr, pearsonr\n", + "from tqdm import tqdm\n", + "\n", + "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + "\n", + "print(\"=\" * 65)\n", + "print(\"BENCHMARK: Consensus Student vs Individual BERTs\")\n", + "print(\"=\" * 65)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# LOAD BENCHMARKS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def load_stsb():\n", + " \"\"\"Load STS-B test set.\"\"\"\n", + " from datasets import load_dataset\n", + " ds = load_dataset(\"mteb/stsbenchmark-sts\", split=\"test\")\n", + " pairs = []\n", + " for row in ds:\n", + " pairs.append({\n", + " \"sent1\": row[\"sentence1\"],\n", + " \"sent2\": row[\"sentence2\"],\n", + " \"score\": row[\"score\"],\n", + " })\n", + " print(f\" STS-B test: {len(pairs)} pairs, scores {min(p['score'] for p in pairs):.1f}-{max(p['score'] for p in pairs):.1f}\")\n", + " return pairs\n", + "\n", + "\n", + "def load_sick():\n", + " \"\"\"Load SICK-R test set.\"\"\"\n", + " from datasets import load_dataset\n", + " ds = load_dataset(\"mteb/sickr-sts\", split=\"test\")\n", + " pairs = []\n", + " for row in ds:\n", + " pairs.append({\n", + " \"sent1\": row[\"sentence1\"],\n", + " \"sent2\": row[\"sentence2\"],\n", + " \"score\": row[\"score\"],\n", + " })\n", + " print(f\" SICK-R test: {len(pairs)} pairs, scores {min(p['score'] for p in pairs):.1f}-{max(p['score'] for p in pairs):.1f}\")\n", + " return pairs\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# ENCODE FUNCTIONS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "@torch.no_grad()\n", + "def encode_with_hf_model(model, tokenizer, texts, batch_size=128, max_len=128):\n", + " \"\"\"Mean-pooled encoding from any HF model.\"\"\"\n", + " all_emb = []\n", + " for i in range(0, len(texts), batch_size):\n", + " batch = texts[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_emb.append(F.normalize(pooled, dim=-1).cpu())\n", + " return torch.cat(all_emb)\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def encode_with_student(student, tokenizer, texts, batch_size=128, max_len=128):\n", + " \"\"\"Encode using the distilled student.\"\"\"\n", + " all_emb = []\n", + " for i in range(0, len(texts), batch_size):\n", + " batch = texts[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " emb = student(inputs[\"input_ids\"], inputs[\"attention_mask\"])\n", + " all_emb.append(emb.cpu())\n", + " return torch.cat(all_emb)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EVALUATION\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def eval_sts(pairs, emb1, emb2):\n", + " \"\"\"Compute Spearman and Pearson correlation on STS-style task.\"\"\"\n", + " cosines = F.cosine_similarity(emb1, emb2, dim=-1).numpy()\n", + " gold = np.array([p[\"score\"] for p in pairs])\n", + " spearman = spearmanr(cosines, gold).statistic\n", + " pearson = pearsonr(cosines, gold).statistic\n", + " return {\n", + " \"spearman\": float(spearman),\n", + " \"pearson\": float(pearson),\n", + " \"cos_mean\": float(cosines.mean()),\n", + " \"cos_std\": float(cosines.std()),\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# STUDENT MODEL (must match training architecture)\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "import torch.nn as nn\n", + "\n", + "class CaptionEncoder(nn.Module):\n", + " def __init__(self, vocab_size=30522, max_len=128, d_model=384,\n", + " n_heads=6, n_layers=6, d_ff=1536, output_dim=768,\n", + " dropout=0.1, pad_token_id=0):\n", + " super().__init__()\n", + " self.pad_token_id = pad_token_id\n", + " self.token_emb = nn.Embedding(vocab_size, d_model, padding_idx=pad_token_id)\n", + " self.pos_emb = nn.Embedding(max_len, d_model)\n", + " self.emb_norm = nn.LayerNorm(d_model)\n", + " self.emb_drop = nn.Dropout(dropout)\n", + " encoder_layer = nn.TransformerEncoderLayer(\n", + " d_model=d_model, nhead=n_heads, dim_feedforward=d_ff,\n", + " dropout=dropout, activation=\"gelu\", batch_first=True,\n", + " norm_first=True)\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n", + " self.output_proj = nn.Sequential(\n", + " nn.Linear(d_model, d_model), nn.GELU(),\n", + " nn.LayerNorm(d_model), nn.Linear(d_model, output_dim))\n", + "\n", + " def forward(self, input_ids, attention_mask=None):\n", + " B, L = input_ids.shape\n", + " positions = torch.arange(L, device=input_ids.device).unsqueeze(0)\n", + " x = self.token_emb(input_ids) + self.pos_emb(positions)\n", + " x = self.emb_drop(self.emb_norm(x))\n", + " if attention_mask is not None:\n", + " kpm = ~attention_mask.bool()\n", + " else:\n", + " kpm = (input_ids == self.pad_token_id)\n", + " x = self.encoder(x, src_key_padding_mask=kpm)\n", + " if attention_mask is not None:\n", + " mask = attention_mask.unsqueeze(-1).float()\n", + " else:\n", + " mask = (~kpm).unsqueeze(-1).float()\n", + " pooled = (x * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " return F.normalize(self.output_proj(pooled), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run_benchmarks():\n", + " from transformers import AutoModel, AutoTokenizer\n", + " import gc\n", + "\n", + " # ── Load benchmarks ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING BENCHMARKS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " stsb = load_stsb()\n", + " sick = load_sick()\n", + "\n", + " stsb_s1 = [p[\"sent1\"] for p in stsb]\n", + " stsb_s2 = [p[\"sent2\"] for p in stsb]\n", + " sick_s1 = [p[\"sent1\"] for p in sick]\n", + " sick_s2 = [p[\"sent2\"] for p in sick]\n", + "\n", + " results = {}\n", + "\n", + " # ── Evaluate student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"EVALUATING: Distilled Consensus Student\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student_tok = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + "\n", + " # Try loading from 200K path first, then 20K\n", + " student = None\n", + " for save_dir in [\"/home/claude/consensus_200k/student\",\n", + " \"/home/claude/distilled_consensus\"]:\n", + " for ckpt in [\"best_model.pt\", \"final_model.pt\"]:\n", + " p = os.path.join(save_dir, ckpt)\n", + " if os.path.exists(p):\n", + " student = CaptionEncoder(\n", + " vocab_size=student_tok.vocab_size,\n", + " max_len=128, d_model=384, n_heads=6, n_layers=6,\n", + " d_ff=1536, output_dim=768, dropout=0.0,\n", + " pad_token_id=student_tok.pad_token_id).to(DEVICE)\n", + " student.load_state_dict(\n", + " torch.load(p, weights_only=True, map_location=DEVICE))\n", + " student.eval()\n", + " n_params = sum(pp.numel() for pp in student.parameters())\n", + " print(f\" Loaded: {p}\")\n", + " print(f\" Parameters: {n_params:,}\")\n", + " break\n", + " if student is not None:\n", + " break\n", + "\n", + " if student is None:\n", + " print(\" ERROR: No student checkpoint found!\")\n", + " return\n", + "\n", + " # Encode\n", + " print(\" Encoding STS-B...\")\n", + " s_stsb1 = encode_with_student(student, student_tok, stsb_s1)\n", + " s_stsb2 = encode_with_student(student, student_tok, stsb_s2)\n", + " print(\" Encoding SICK-R...\")\n", + " s_sick1 = encode_with_student(student, student_tok, sick_s1)\n", + " s_sick2 = encode_with_student(student, student_tok, sick_s2)\n", + "\n", + " r_stsb = eval_sts(stsb, s_stsb1, s_stsb2)\n", + " r_sick = eval_sts(sick, s_sick1, s_sick2)\n", + " results[\"student\"] = {\"stsb\": r_stsb, \"sick\": r_sick, \"params\": n_params}\n", + " print(f\" STS-B: spearman={r_stsb['spearman']:.4f} pearson={r_stsb['pearson']:.4f}\")\n", + " print(f\" SICK-R: spearman={r_sick['spearman']:.4f} pearson={r_sick['pearson']:.4f}\")\n", + "\n", + " del student\n", + " gc.collect()\n", + " torch.cuda.empty_cache()\n", + "\n", + " # ── Evaluate individual BERTs ──\n", + " bert_models = [\n", + " (\"google-bert/bert-base-uncased\", \"bert-base\"),\n", + " (\"answerdotai/ModernBERT-base\", \"modern-bert\"),\n", + " (\"FacebookAI/roberta-base\", \"roberta\"),\n", + " (\"albert/albert-base-v2\", \"albert\"),\n", + " (\"distilbert/distilbert-base-uncased\", \"distilbert\"),\n", + " ]\n", + "\n", + " for model_name, short_name in bert_models:\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"EVALUATING: {short_name} ({model_name})\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " n_p = sum(p.numel() for p in model.parameters())\n", + " print(f\" Parameters: {n_p:,}\")\n", + "\n", + " print(\" Encoding STS-B...\")\n", + " e_stsb1 = encode_with_hf_model(model, tokenizer, stsb_s1)\n", + " e_stsb2 = encode_with_hf_model(model, tokenizer, stsb_s2)\n", + " print(\" Encoding SICK-R...\")\n", + " e_sick1 = encode_with_hf_model(model, tokenizer, sick_s1)\n", + " e_sick2 = encode_with_hf_model(model, tokenizer, sick_s2)\n", + "\n", + " r_stsb = eval_sts(stsb, e_stsb1, e_stsb2)\n", + " r_sick = eval_sts(sick, e_sick1, e_sick2)\n", + " results[short_name] = {\"stsb\": r_stsb, \"sick\": r_sick, \"params\": n_p}\n", + " print(f\" STS-B: spearman={r_stsb['spearman']:.4f} pearson={r_stsb['pearson']:.4f}\")\n", + " print(f\" SICK-R: spearman={r_sick['spearman']:.4f} pearson={r_sick['pearson']:.4f}\")\n", + "\n", + " del model\n", + " gc.collect()\n", + " torch.cuda.empty_cache()\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # SUMMARY\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\"\\n {'Model':<20} {'Params':>12} {'STS-B ρ':>10} {'SICK-R ρ':>10}\")\n", + " print(f\" {'-'*52}\")\n", + "\n", + " # Sort by STS-B spearman\n", + " sorted_results = sorted(results.items(),\n", + " key=lambda x: x[1][\"stsb\"][\"spearman\"], reverse=True)\n", + " for name, r in sorted_results:\n", + " marker = \" ★\" if name == \"student\" else \"\"\n", + " print(f\" {name:<20} {r['params']:>10,} \"\n", + " f\"{r['stsb']['spearman']:>10.4f} {r['sick']['spearman']:>10.4f}{marker}\")\n", + "\n", + " # Student vs best individual\n", + " student_stsb = results[\"student\"][\"stsb\"][\"spearman\"]\n", + " best_name = max((n for n in results if n != \"student\"),\n", + " key=lambda n: results[n][\"stsb\"][\"spearman\"])\n", + " best_stsb = results[best_name][\"stsb\"][\"spearman\"]\n", + " best_params = results[best_name][\"params\"]\n", + " student_params = results[\"student\"][\"params\"]\n", + "\n", + " print(f\"\\n Student STS-B: {student_stsb:.4f} ({student_params:,} params)\")\n", + " print(f\" Best teacher: {best_stsb:.4f} ({best_name}, {best_params:,} params)\")\n", + " print(f\" Gap: {student_stsb - best_stsb:+.4f}\")\n", + " print(f\" Param ratio: {best_params / student_params:.1f}×\")\n", + "\n", + " # Save\n", + " save_path = \"/home/claude/benchmark_results.json\"\n", + " with open(save_path, \"w\") as f:\n", + " json.dump(results, f, indent=2, default=str)\n", + " print(f\"\\n Saved to {save_path}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run_benchmarks()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "445251c2934740f2bf149cd2e7848887", + "94548dbad7fd4c8386087683768b5644", + "8654b6b67b294522a6c6005820745747", + "2da39e52fb8d463bae47fdb295fd95c2", + "fab849e9e4d34fdcb47fb6f5b37ce8b7", + "e71d4fe04ba14e41b3d34012299b33f0", + "24ea33a9dd494980902f2d2d2d3c2f2e", + "f993dc3a96564c188d3cade339ffaf7e", + "4b7bde4ca12d4d3c99d1fd7a3efe1dc9", + "6ce03d413a0849a78aa4683cbd3333d6", + "8fe8bf9594dd451d94f976ae77c6c309", + "b77c9510ee784585aa2856f8c4e0787f", + "c7e1fe16b5614caf859d30fa44db4be9", + "39773bab82b241fbb2999844c9812516", + "e88a3f9b46a64d2f9f535cd9eb713c70", + "dcfd19088e03463990d21307614ccd6a", + "441822419f114790abbdea3fe0f87d0d", + "4197c1a8df1d45999669aba6e088fe83", + "2da61cc4d3d24578911ad10a79fd98d0", + "cd89f0c06e7c408aa15df2668eb89c2b", + "5914712138f840d4830a08e1286818bd", + "8b01634426a646a58a4bb4fff708ea9d", + "a3ec1bca46b54d608ed1cec3e9b86a77", + "e373cfc9515b414893098c943e9b3544", + "7922153899924061b12e434ffb84d37a", + "feb0891d370a4ab0b617e4ff7c9fe7b8", + "44cb8dc7bbce44c49baf4d5239b6bb25", + "33165e1fd85f4d33882db63d0d59cae4", + "059a83e42aab417a8e3b451d06ecbfe8", + "c9963fa14058479b83ba8fd96addd7d9", + "7a9f11ff114e4093a2b03366c2273164", + "ce79ca6d9e864136be126a1474c8d389", + "35d1c561a6c145d183f24daf78c71453", + "b1f74d52b7594a8db5b1ede3cb901715", + "3e897c0e4f804c1bac03d892bbfe5022", + "2277e5a2cd184f928daeb73c0acbde94", + "43cc92e2c6d245d097ab3d15f3e77f2e", + "e818f395a11d4fdeaa03a54bddabc91c", + "6f644b61c77b480a868b391ee704d51c", + "1b35574690a3408eb527b43c6775bbd4", + "90cc3d74316f4c59973a067b035e3d20", + "0cd89d5a5abc4654965285254ad615da", + "358642b7ed2a43868cc8b702ce0efc6c", + "adf03a58a46749b7bae27df5f5b018ed", + "79f9be2aa6964b42b4eafe00c3856d35", + "8ffcc3129027449b827a5f3edfad279d", + "0a05d9c36747437e9a1962213fdf328e", + "c65bc22e7c654ac2ba342f548ba26344", + "5f77c2218bfe4af9b27e457c3dd59c9a", + "282bd748e6874a78ac59f284a9603612", + "fd695faf735d487f92bb9e22be548f6c", + "adbda2deb81e4ea68089a699d1e11a4d", + "4a06241efad54e4d9ffaf09c8ef1b391", + "f4b6dcb6d04f42d4b06641f2b3c7fa72", + "68393608a5d74bd9879581418ef36f4c", + "bebc7953dede45b9b483fcebd95d25a7", + "4ee6f2b4b6574d43a3e8eebe8a71e64b", + "2408085f6d0246b690f48e81ddf5918a", + "eeae78b6686a4264b0d220c0682bcb09", + "eb74e67411924f15ab8d1bef2af9dcce", + "edb8811d4349428da14d5fc849c5e6d9", + "e9594b96358a4a88a3ddd737ff7ed646", + "11d49bd045244030abde1edf67fd175f", + "bf00077cc1e6479595ea23856fb2b857", + "7790be0e03fc4313bd9a2db205c933fa", + "b3ca1995e07548608b38e2dc9cd58b4a", + "2db067799aaa458390f2b4f6512b1648", + "c5ea028e856e49499013b6247fbae032", + "0a9708fd8cc3484bac185ce1e3363053", + "8299c7f71a724245aac146136ce68f14", + "b7e69e58d28842d389ec3af6efc225da", + "eaeaab1835084444abc4ee5d175be185", + "721d0a2d54d44806b465471a40e65e48", + "f66edb1067f6413ba8f954a4b1455f53", + "7ab650307580439aa871ed868dd73e02", + "b08b134a4a654ce1b58bade830939e43", + "db5e9dbf211f482fb729859345411acb", + "cfc335914cbf43b084f2884bfcd16b08", + "6ddcd1a0106b43898e9bfe457b737b50", + "2a85e704dc5042a783295611891d3673", + "e4d4c1dae6874d54b99f92f89bca7318", + "f5b69ff3b2a14f4e8b44b012a72fae5b", + "ada3af63f20e46a59ab3d496f477610d", + "24c77bb630ab4c159a75647090b618de", + "20550b2196d64a129e9d45107395ae25", + "c6a0987845784bdb8ae012cf8884769a", + "76fd271429ac46ab8418a8ac6ffc0cb9", + "764233414e36462f95f94269c63d710b", + "189700333aa048728934be7b5d828645", + "30b4705e2bf045f9beac5635eada4e82", + "a414d6711a94476b995ab71e5fb5fccc", + "689bee8e0e964a4a853af9a3e00bb2f0", + "a52a7381cfdb45d59f2c8d464ecfe4bf", + "60aa1ad490ec453abac816adbecad28e", + "0ccfd37c64df4fc2bc9c2019acb4f3fb", + "38b077de9ea74eb6befdca50b10f4a8e", + "221b0c6322624ee8817b9c1d35315b73", + "66c3da07e60e4ea4a86739687596620c", + "dccd8c3c10a34ff9a6de3870ee0645f8", + "e0a5762e6fdd486a897fa94109e3b988", + "d9389ca3df834d2f9a64e8b0bb308a29", + "aeece63f05df4ce78953275473abdbad", + "0b4bc34ed08e4a24b28c280cc3063646", + "bdd1e56b56cb4079a5c6395c825daf72", + "c667166f08174334b209b0be75da36f8", + "7eab67765b894fb1883704e73df5b6d0", + "ab9914baaa1a4481b26c702d112b2d5a", + "c3f947b3292c457f91f8591b35b2564a", + "046cf8d05fe444589e58d255dbf5db8f", + "f5c305fa31ce443c940a9f16c1ef68ae", + "508568909bae475b9c29be12d70b0ccd", + "b7bd72832268451f80619571231406e6", + "4e7a916f70064f319c23928c44c1b3de", + "e7e7ee73b9e743f8973fb3ba937d78dc", + "4462cb5698df46e18cdfccba1f7538f9", + "44b68e0e5de24b5f80043a9429588b66", + "1c38a351b71e49f580822113040568cc", + "a11f1eb907c74990a36234b1dc0970ea", + "e3b66d14d74048e9ae0d215148121b6b", + "bd5723c7ac0b4211b32d76b8b775f14f", + "9092d99755a347ac8a45ae016e819ee7", + "7f618a7728e948168acd1d46a016022f", + "e9317d14d03d47b58112946964fccba4", + "c3d63defe838429892e07e3fe4743525", + "3c01fb2c7fe947838b39254ef68096bb", + "bde8ae428f394b2e97af1a09d1171612", + "960f59cc25f94d38b47214d9958d9eb0", + "eb6b2d96790243ee9bc6bd581f1c0d7b", + "afb1d381e40e4bc7b96bb3e05bd5b01e", + "79a83cd59ee04936b90d2106330af650", + "e9c2f24d106a43bbaa576b910f7b10a2", + "8ebfbf0823b44c668ec5db96e177a7f2", + "925475ff7a5a456690873c59a688d1cf", + "35e71e1fecbb4c5db7d4848056cb273e", + "44ca9db110fa465bb6aea334fae9fb73", + "eb399846160a4e5a9ac1fb44f01c554a", + "01e5df033e78448c8fea2a10c601eed5", + "8b73d9c5065741cf930497f238fec29c", + "d2f8cc33ff5b4ed9b183c7a80bfbf17c", + "8ab7372bef074e47bc6620bb94949a19", + "5f725231dd86474aa60ac6fde76508b7", + "93a148c2568943d89dde0d3a9ad9aece", + "2733ae5d3ae049f0bff7929ee3f256fa", + "41235a1f501c4280a21a96b6a0415a35", + "fe8ac8b38d5949bf957b0789371af8eb", + "1f6c77432fb74833abecca6146cb2ccf", + "022ed8d5166142b4a5a97d0df26e8048", + "bfb3b36c2947429cb7474628ed89e573", + "fe6758b05193460bac4f73e05a6c4f5f", + "ceb23d6435844a3eb4beaab9e103eca6", + "256955cf07f04930b76d2c7bcef293ac", + "e4afa7b6702b486c8531f2f5c0e1821e", + "ebde16de147546f0918466c31897aef5", + "d5d72d42cddc48a287b7076758543c95", + "6dbac12dff1d4d82a8abf2c295eb7788", + "a4967e28ec7b4203947aff3d3f98ab5f", + "1ab19f3149f341fdb27fb6b2fcd5cbf1", + "b267a673008045d88549e9248bfd3a64", + "19cfe604fa5c4d3db43f0a32e8a35c42", + "ab66728ec8eb47b8923f85535a34a8f5", + "a581ce9c6d684f5d9dca44b18f62a034", + "7baa5ad8efe44571a33999119a9a4b5d", + "ecbb8b12316c42efb6dec051b55bc1fa", + "ca2a86d355644fdcb204a44ced3d4ec7", + "d2fa9d75def7493b8878733b386e076c" + ] + }, + "id": "zWMH2My6rNUw", + "outputId": "bc119d38-897f-4ba2-bf9c-eea93134528b" + }, + "execution_count": 9, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "BENCHMARK: Consensus Student vs Individual BERTs\n", + "=================================================================\n", + "\n", + "=================================================================\n", + "LOADING BENCHMARKS\n", + "=================================================================\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "README.md: 0.00B [00:00, ?B/s]" + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "445251c2934740f2bf149cd2e7848887" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "train.jsonl.gz: 0%| | 0.00/278k [00:00 0:\n", + " vols.append(v)\n", + " if len(vols) < 10:\n", + " return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# DATA: Caption loading + 5-BERT extraction\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def load_captions(n, min_len=50):\n", + " from datasets import load_dataset\n", + " print(f\"\\n Loading captions (n={n:,})...\")\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > min_len:\n", + " captions.append(cap)\n", + " if len(captions) >= n:\n", + " break\n", + " print(f\" Got {len(captions):,} captions\")\n", + " return captions\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def extract_one(model_name, short_name, captions, max_len, batch_size):\n", + " \"\"\"Extract mean-pooled embeddings from a single HF model.\"\"\"\n", + " from transformers import AutoModel, AutoTokenizer\n", + " print(f\"\\n Extracting: {short_name} ({model_name}, max_len={max_len})...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " dim = model.config.hidden_size\n", + " n_params = sum(p.numel() for p in model.parameters())\n", + " print(f\" dim={dim}, {n_params:,} params\")\n", + "\n", + " all_emb = []\n", + " for i in tqdm(range(0, len(captions), batch_size), desc=f\" {short_name}\"):\n", + " batch = captions[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + "\n", + " emb = torch.cat(all_emb)\n", + " print(f\" Shape: {emb.shape}\")\n", + " del model, tokenizer\n", + " gc.collect()\n", + " torch.cuda.empty_cache()\n", + " return emb\n", + "\n", + "\n", + "def extract_all():\n", + " \"\"\"Extract embeddings from all 5 models. Caches to disk.\"\"\"\n", + " os.makedirs(CFG.cache_dir, exist_ok=True)\n", + " caps_path = os.path.join(CFG.cache_dir, \"captions.json\")\n", + "\n", + " all_cached = all(\n", + " os.path.exists(os.path.join(CFG.cache_dir, f\"{s}.pt\"))\n", + " for _, s, _ in MODELS)\n", + "\n", + " if all_cached:\n", + " print(\"\\n Loading cached embeddings...\")\n", + " embeds = {}\n", + " for _, short, _ in MODELS:\n", + " embeds[short] = torch.load(\n", + " os.path.join(CFG.cache_dir, f\"{short}.pt\"), weights_only=True)\n", + " print(f\" {short}: {embeds[short].shape}\")\n", + "\n", + " # Load or regenerate captions\n", + " if os.path.exists(caps_path):\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " print(f\" Captions loaded: {len(captions):,}\")\n", + " else:\n", + " print(\" captions.json missing, regenerating...\")\n", + " captions = load_captions(CFG.n_samples, CFG.min_caption_len)\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " # Trim to smallest common size\n", + " n = min(len(captions), min(e.shape[0] for e in embeds.values()))\n", + " captions = captions[:n]\n", + " embeds = {k: v[:n] for k, v in embeds.items()}\n", + " print(f\" Using {n:,} samples\")\n", + " return embeds, captions\n", + "\n", + " captions = load_captions(CFG.n_samples, CFG.min_caption_len)\n", + "\n", + " embeds = {}\n", + " for model_name, short, model_max_len in MODELS:\n", + " out_path = os.path.join(CFG.cache_dir, f\"{short}.pt\")\n", + " if os.path.exists(out_path):\n", + " print(f\"\\n {short}: cached, loading...\")\n", + " embeds[short] = torch.load(out_path, weights_only=True)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " continue\n", + " emb = extract_one(model_name, short, captions,\n", + " model_max_len, CFG.extract_batch)\n", + " # Ensure 768-dim\n", + " if emb.shape[1] != 768:\n", + " if emb.shape[1] < 768:\n", + " emb = F.pad(emb, (0, 768 - emb.shape[1]))\n", + " else:\n", + " emb = emb[:, :768]\n", + " embeds[short] = emb\n", + " torch.save(emb, os.path.join(CFG.cache_dir, f\"{short}.pt\"))\n", + "\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " return embeds, captions\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# WHITENED PROCRUSTES ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " \"\"\"Covariance matrix inverse square root for whitening.\"\"\"\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "\n", + "def procrustes_align(source, target, n_align=10000):\n", + " \"\"\"Whitened Procrustes: center → whiten → normalize → SVD rotate.\"\"\"\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean\n", + " Tc = T - t_mean\n", + " N_s = Sc.shape[0]\n", + "\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + "\n", + " # Whiten\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + "\n", + " # SVD rotation\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + "\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " t_unwhiten = torch.linalg.pinv(t_whiten)\n", + "\n", + " return {\n", + " \"rotation\": R,\n", + " \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": t_unwhiten,\n", + " \"cos_before\": cos_before,\n", + " \"cos_after\": cos_after,\n", + " }\n", + "\n", + "\n", + "def apply_align(emb, a):\n", + " \"\"\"Apply whitened Procrustes: center → whiten → rotate → unwhiten.\"\"\"\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]\n", + " x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]\n", + " return x\n", + "\n", + "\n", + "def generate_consensus(embeds):\n", + " \"\"\"\n", + " Align all 5 models to shared space, return normalized centroid.\n", + " The five-BERT pentachoron experiment proved this centroid is a\n", + " geometric constant (identical to 3 decimal places across 5 seeds).\n", + " No learned model needed.\n", + " \"\"\"\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"WHITENED PROCRUSTES ALIGNMENT + CONSENSUS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref_name = \"bert\"\n", + " names = [s for _, s, _ in MODELS]\n", + " aligned = {}\n", + "\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], embeds[ref_name])\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " label = \" (ref)\" if name == ref_name else \"\"\n", + " print(f\" {name:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}{label}\")\n", + "\n", + " # Consensus = normalized centroid\n", + " centroid = sum(aligned[n] for n in names) / len(names)\n", + " consensus = F.normalize(centroid, dim=-1)\n", + "\n", + " # Verify\n", + " N_check = min(5000, consensus.shape[0])\n", + " for name in names:\n", + " cos = F.cosine_similarity(\n", + " consensus[:N_check], aligned[name][:N_check], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name:10s}): {cos:.4f}\")\n", + "\n", + " return consensus\n", + "\n", + "\n", + "# ═══════════════════════════════════════════════════════════���══════\n", + "# TRAINING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def train():\n", + " torch.manual_seed(CFG.seed)\n", + " torch.cuda.manual_seed_all(CFG.seed)\n", + " np.random.seed(CFG.seed)\n", + "\n", + " # ── Extract + Align + Consensus ──\n", + " embeds, captions = extract_all()\n", + " consensus = generate_consensus(embeds)\n", + "\n", + " # Free raw embeddings\n", + " del embeds\n", + " gc.collect()\n", + " torch.cuda.empty_cache()\n", + "\n", + " # ── Tokenize ──\n", + " from transformers import AutoTokenizer\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " print(f\"\\n Tokenizer: bert-base-uncased (vocab={tokenizer.vocab_size})\")\n", + "\n", + " print(\" Pre-tokenizing...\")\n", + " all_ids, all_masks = [], []\n", + " chunk = 50000\n", + " for i in tqdm(range(0, len(captions), chunk), desc=\" Tokenizing\"):\n", + " j = min(i + chunk, len(captions))\n", + " tokens = tokenizer(captions[i:j], max_length=CFG.tokenize_len,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\")\n", + " all_ids.append(tokens[\"input_ids\"])\n", + " all_masks.append(tokens[\"attention_mask\"])\n", + "\n", + " input_ids = torch.cat(all_ids)\n", + " attention_mask = torch.cat(all_masks)\n", + "\n", + " real_lens = attention_mask.sum(1).float()\n", + " print(f\" Token lengths: mean={real_lens.mean():.0f} \"\n", + " f\"median={real_lens.median():.0f} \"\n", + " f\">{CFG.tokenize_len}: {(real_lens >= CFG.tokenize_len).float().mean():.1%}\")\n", + " print(f\" Padded to: {CFG.tokenize_len} (model supports up to {CFG.max_len})\")\n", + "\n", + " # Split\n", + " n_train = len(captions) - CFG.n_val\n", + " print(f\" Train: {n_train:,}, Val: {CFG.n_val:,}\")\n", + "\n", + " # Move to GPU\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " # ── Build student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"STUDENT MODEL: CaptionEncoder\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student = CaptionEncoder(\n", + " vocab_size=tokenizer.vocab_size,\n", + " max_len=CFG.max_len,\n", + " d_model=CFG.d_model,\n", + " n_heads=CFG.n_heads,\n", + " n_layers=CFG.n_layers,\n", + " d_ff=CFG.d_ff,\n", + " output_dim=CFG.output_dim,\n", + " dropout=CFG.dropout,\n", + " pad_token_id=tokenizer.pad_token_id,\n", + " ).to(DEVICE)\n", + "\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " size_mb = sum(p.numel() * p.element_size() for p in student.parameters()) / 1e6\n", + " print(f\" Architecture: {CFG.n_layers}L, {CFG.d_model}d, {CFG.n_heads}h, {CFG.d_ff} FFN\")\n", + " print(f\" Position capacity: {CFG.max_len}\")\n", + " print(f\" Output: {CFG.output_dim}-dim (consensus space)\")\n", + " print(f\" Parameters: {n_params:,}\")\n", + " print(f\" Size: {size_mb:.1f} MB\")\n", + "\n", + " # ── Warm-start from previous checkpoint if available ──\n", + " warm_started = False\n", + " for prev_dir in [\"/home/claude/consensus_200k/student\",\n", + " \"/home/claude/distilled_consensus\"]:\n", + " prev_ckpt = os.path.join(prev_dir, \"best_model.pt\")\n", + " if os.path.exists(prev_ckpt):\n", + " print(f\"\\n Warm-starting from: {prev_ckpt}\")\n", + " prev_state = torch.load(prev_ckpt, weights_only=True, map_location=DEVICE)\n", + " current_state = student.state_dict()\n", + "\n", + " loaded, extended, skipped = 0, 0, 0\n", + " for name, param in prev_state.items():\n", + " if name not in current_state:\n", + " skipped += 1\n", + " continue\n", + " if param.shape == current_state[name].shape:\n", + " current_state[name] = param\n", + " loaded += 1\n", + " elif \"pos_emb\" in name and param.shape[0] < current_state[name].shape[0]:\n", + " # Extend position embeddings: copy old, init new\n", + " old_len = param.shape[0]\n", + " current_state[name][:old_len] = param\n", + " nn.init.normal_(current_state[name][old_len:], std=0.02)\n", + " extended += 1\n", + " print(f\" Extended {name}: {param.shape[0]}→{current_state[name].shape[0]}\")\n", + " else:\n", + " skipped += 1\n", + " print(f\" Skipped {name}: {param.shape}→{current_state[name].shape}\")\n", + "\n", + " student.load_state_dict(current_state)\n", + " print(f\" Loaded: {loaded}, Extended: {extended}, Skipped: {skipped}\")\n", + " warm_started = True\n", + " break\n", + "\n", + " if not warm_started:\n", + " print(\"\\n Training from scratch (no previous checkpoint found)\")\n", + "\n", + " # ── Optimizer + Scheduler ──\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=CFG.lr,\n", + " weight_decay=CFG.weight_decay)\n", + " n_batches = n_train // CFG.batch_size\n", + " total_steps = n_batches * CFG.epochs\n", + " scheduler = torch.optim.lr_scheduler.SequentialLR(\n", + " optimizer,\n", + " [torch.optim.lr_scheduler.LinearLR(optimizer, start_factor=0.01,\n", + " total_iters=CFG.warmup_steps),\n", + " torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=max(total_steps - CFG.warmup_steps, 1),\n", + " eta_min=1e-6)],\n", + " milestones=[CFG.warmup_steps])\n", + "\n", + " save_dir = os.path.join(CFG.cache_dir, \"student\")\n", + " os.makedirs(save_dir, exist_ok=True)\n", + "\n", + " # ── Training loop ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({CFG.epochs} epochs, {n_batches} batches/epoch)\")\n", + " print(f\" Losses: InfoNCE + MSE + pentachoron CV (target={CFG.cv_target})\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " all_metrics = {\n", + " \"config\": {k: str(v) for k, v in vars(CFG).items()},\n", + " \"warm_started\": warm_started,\n", + " \"epochs\": [],\n", + " }\n", + " best_val_cos = 0.0\n", + "\n", + " for epoch in range(CFG.epochs):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " losses = {\"total\": 0, \"nce\": 0, \"mse\": 0}\n", + " metrics = {\"acc\": 0, \"cos\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, CFG.batch_size):\n", + " idx = perm[i:i+CFG.batch_size]\n", + " if len(idx) < 8:\n", + " continue\n", + "\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + "\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=CFG.cv_target)\n", + "\n", + " loss = (CFG.nce_weight * l_nce +\n", + " CFG.mse_weight * l_mse +\n", + " CFG.cv_weight * l_cv)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), CFG.grad_clip)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + "\n", + " losses[\"total\"] += loss.item()\n", + " losses[\"nce\"] += l_nce.item()\n", + " losses[\"mse\"] += l_mse.item()\n", + " metrics[\"acc\"] += acc\n", + " metrics[\"cos\"] += cos\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Validation\n", + " student.eval()\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, CFG.n_val, 512):\n", + " vj = min(vi + 512, CFG.n_val)\n", + " ve = student(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(ve)\n", + " val_emb = torch.cat(val_embs)\n", + " _, val_acc = infonce(val_emb[:2000], val_targets[:2000])\n", + " val_cos = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " val_cv = cv_metric(val_emb[:2000])\n", + "\n", + " summary = {\n", + " \"epoch\": epoch + 1, \"elapsed\": elapsed,\n", + " \"loss\": losses[\"total\"] / d,\n", + " \"train_acc\": metrics[\"acc\"] / d,\n", + " \"train_cos\": metrics[\"cos\"] / d,\n", + " \"val_acc\": val_acc, \"val_cos\": val_cos, \"val_cv\": val_cv,\n", + " }\n", + " all_metrics[\"epochs\"].append(summary)\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s \"\n", + " f\"loss={summary['loss']:.4f} \"\n", + " f\"t_acc={summary['train_acc']:.3f} t_cos={summary['train_cos']:.3f} \"\n", + " f\"v_acc={summary['val_acc']:.3f} v_cos={summary['val_cos']:.3f} \"\n", + " f\"v_cv={summary['val_cv']:.3f}\")\n", + "\n", + " # Save best\n", + " if val_cos > best_val_cos:\n", + " best_val_cos = val_cos\n", + " torch.save(student.state_dict(), os.path.join(save_dir, \"best_model.pt\"))\n", + "\n", + " # Periodic save\n", + " if (epoch + 1) % 10 == 0:\n", + " torch.save(student.state_dict(),\n", + " os.path.join(save_dir, f\"model_e{epoch+1:02d}.pt\"))\n", + "\n", + " # Final save\n", + " torch.save(student.state_dict(), os.path.join(save_dir, \"final_model.pt\"))\n", + " tokenizer.save_pretrained(os.path.join(save_dir, \"tokenizer\"))\n", + " with open(os.path.join(save_dir, \"metrics.json\"), \"w\") as f:\n", + " json.dump(all_metrics, f, indent=2, default=str)\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # FINAL EVALUATION\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FINAL EVALUATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.load_state_dict(\n", + " torch.load(os.path.join(save_dir, \"best_model.pt\"),\n", + " weights_only=True, map_location=DEVICE))\n", + " student.eval()\n", + "\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, CFG.n_val, 512):\n", + " vj = min(vi + 512, CFG.n_val)\n", + " ve = student(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(ve)\n", + " val_emb = torch.cat(val_embs)\n", + "\n", + " # Retrieval (on 2K subset)\n", + " sub = min(2000, CFG.n_val)\n", + " sim = val_emb[:sub] @ val_targets[:sub].T\n", + " labels = torch.arange(sub, device=DEVICE)\n", + " r1 = (sim.argmax(1) == labels).float().mean().item()\n", + " r5 = (sim.topk(5, dim=1).indices == labels.unsqueeze(1)).any(1).float().mean().item()\n", + " r10 = (sim.topk(10, dim=1).indices == labels.unsqueeze(1)).any(1).float().mean().item()\n", + "\n", + " cos_match = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " final_cv = cv_metric(val_emb[:2000])\n", + "\n", + " print(f\" Retrieval (student → consensus):\")\n", + " print(f\" R@1: {r1:.4f}\")\n", + " print(f\" R@5: {r5:.4f}\")\n", + " print(f\" R@10: {r10:.4f}\")\n", + " print(f\" Cosine: {cos_match:.4f}\")\n", + " print(f\" CV: {final_cv:.4f} (target: {CFG.cv_target})\")\n", + " print(f\" Model: {n_params:,} params, {size_mb:.1f} MB\")\n", + "\n", + " # Standalone inference test\n", + " print(f\"\\n Standalone similarity test:\")\n", + " test = [\n", + " \"A cat sitting on a windowsill watching birds outside\",\n", + " \"A golden retriever playing fetch on the beach at sunset\",\n", + " \"A still life painting with flowers and fruit on a table\",\n", + " \"An aerial photograph of a city skyline at night\",\n", + " \"A child riding a bicycle through autumn leaves in a park\",\n", + " ]\n", + " with torch.no_grad():\n", + " tok = tokenizer(test, max_length=CFG.tokenize_len, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " embs = student(tok[\"input_ids\"], tok[\"attention_mask\"])\n", + " sim = embs @ embs.T\n", + " for i in range(len(test)):\n", + " for j in range(i+1, len(test)):\n", + " print(f\" [{i}]↔[{j}]: {sim[i,j]:.3f} \"\n", + " f\"({test[i][:35]}↔{test[j][:35]})\")\n", + "\n", + " print(f\"\\n Saved to: {save_dir}/\")\n", + " print(f\" Files: best_model.pt, final_model.pt, tokenizer/, metrics.json\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " train()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "3dcbb662b30f47cea1f426e9dc400ccd", + "95e7bd361fd64d5da9eba70f738098f1", + "aed55b8a33774489858846e8d91df294", + "cb2850b0ed5c446da78ae62c66f31ab8", + "177c8cace36648309828ff022da377c6", + "18644eff9b28496f8272ea19fc3b7321", + "f0d53e94e0754085b05a2385bda50e10", + "407aa645322c43ebb807b1042381e460", + "27987162e66c4fd5be03e00881c1767c", + "1502c7cc58e84ee1a9f52e75a9b9b756", + "b41d4555acca4842bcbcb68fedf14b82", + "d80a44e073ac4e4d8a080144d253d7fe", + "5e99b7b4abd448f6b52092186dbaa05e", + "37f6bb1b4c384556b7b21667904d38ba", + "0ca39f5cffca45e9b49e99c5547ff5b6", + "08ff8a06c91646d699bb822e2128e6c4", + "3361780a16294b8aadaf017d28360fd5", + "8038f4a50f4340489f7c367f33a9dc88", + "16f31db48da24f1ea6b6707475a0ba76", + "d8e24943261740ef9b010859cc3fcbc1", + "e72a0689679f4bddbfaad6387d290520", + "b9dd65320fb24523b8bc0159e0f22eac", + "d3e519965795424c8e03db777192988f", + "b2178870751745ae8b2ff59efb912a6a", + "596bca946934466ebac67b58bb55a2c6", + "7239062a47784c0fb07e98e430e69c6c", + "4ce5209809b34b5595c16501acaf36e2", + "a4038ba6cc2745e9817d5442adb12713", + "e0dec83560634bd7bac5bbfcaecb4857", + "4261972c71334de188c80f166b653f15", + "10da6bb464ca419eb17496580bf7d88f", + "7b8ed631bda04a3dbfaeea4870a2a4b1", + "9e729736efbd42218c3775334ae2238e", + "db8fb5f7d5d641e0a63c27883875fbb8", + "359bf897b2404f4ca3b1250d8d060c8b", + "21233148c885454d9d1ccc5467b0877c", + "b8a4c7d2168743028374e094fdba3778", + "e001c3cb8a1a44cca95d5d191749e596", + "49fb49ff55c14127b11e942667fddc5a", + "251a783db5f74b50a7b32970c85fac1a", + "bb23409acd204b52b03879633a2a3f81", + "684d665814904e2ba054607b97a57b55", + "ff5545a913c140ff96f42e3334942bdc", + "a780ba2ef7834be6882edccace4cbe3b" + ] + }, + "id": "N0WAE-pitGyM", + "outputId": "6ad26487-f437-4aba-a20f-2ae0a7b8675f" + }, + "execution_count": 1, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "DISTILLED CONSENSUS BERT — 500K / 8192 pos\n", + "=================================================================\n", + " Device: cuda\n", + " Samples: 500,000\n", + " Student: 6L 384d → 768d\n", + " Pos capacity: 8192, Tokenize pad: 512\n", + "\n", + " Loading captions (n=500,000)...\n", + " Got 500,000 captions\n", + "\n", + " bert: cached, loading...\n", + " Shape: torch.Size([500000, 768])\n", + "\n", + " Extracting: modern (answerdotai/ModernBERT-base, max_len=8192)...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/134 [00:00512: 0.0%\n", + " Padded to: 512 (model supports up to 8192)\n", + " Train: 495,000, Val: 5,000\n", + "\n", + "=================================================================\n", + "STUDENT MODEL: CaptionEncoder\n", + "=================================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/tmp/ipykernel_49962/1732017288.py:120: UserWarning: enable_nested_tensor is True, but self.use_nested_tensor is False because encoder_layer.norm_first was True\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " Architecture: 6L, 384d, 6h, 1536 FFN\n", + " Position capacity: 8192\n", + " Output: 768-dim (consensus space)\n", + " Parameters: 25,958,016\n", + " Size: 103.8 MB\n", + "\n", + " Warm-starting from: /home/claude/consensus_200k/student/best_model.pt\n", + " Extended pos_emb.weight: 128→8192\n", + " Loaded: 81, Extended: 1, Skipped: 0\n", + "\n", + "=================================================================\n", + "TRAINING (30 epochs, 3867 batches/epoch)\n", + " Losses: InfoNCE + MSE + pentachoron CV (target=0.084)\n", + "=================================================================\n", + " E 1: 689s loss=0.0197 t_acc=1.000 t_cos=0.804 v_acc=1.000 v_cos=0.803 v_cv=0.104\n", + " E 2: 688s loss=0.0188 t_acc=1.000 t_cos=0.807 v_acc=1.000 v_cos=0.810 v_cv=0.085\n", + " E 3: 688s loss=0.0176 t_acc=1.000 t_cos=0.811 v_acc=1.000 v_cos=0.820 v_cv=0.103\n", + " E 4: 689s loss=0.0166 t_acc=1.000 t_cos=0.815 v_acc=1.000 v_cos=0.825 v_cv=0.084\n", + " E 5: 689s loss=0.0159 t_acc=1.000 t_cos=0.819 v_acc=1.000 v_cos=0.819 v_cv=0.086\n", + " E 6: 689s loss=0.0153 t_acc=1.000 t_cos=0.821 v_acc=1.000 v_cos=0.821 v_cv=0.095\n", + " E 7: 688s loss=0.0148 t_acc=1.000 t_cos=0.824 v_acc=1.000 v_cos=0.820 v_cv=0.091\n", + " E 8: 689s loss=0.0143 t_acc=1.000 t_cos=0.827 v_acc=1.000 v_cos=0.834 v_cv=0.088\n", + " E 9: 688s loss=0.0139 t_acc=1.000 t_cos=0.829 v_acc=1.000 v_cos=0.829 v_cv=0.088\n", + " E10: 689s loss=0.0136 t_acc=1.000 t_cos=0.831 v_acc=1.000 v_cos=0.829 v_cv=0.087\n", + " E11: 689s loss=0.0133 t_acc=1.000 t_cos=0.833 v_acc=1.000 v_cos=0.836 v_cv=0.082\n", + " E12: 689s loss=0.0129 t_acc=1.000 t_cos=0.835 v_acc=1.000 v_cos=0.838 v_cv=0.084\n", + " E13: 688s loss=0.0126 t_acc=1.000 t_cos=0.837 v_acc=1.000 v_cos=0.842 v_cv=0.083\n", + " E14: 689s loss=0.0123 t_acc=1.000 t_cos=0.839 v_acc=1.000 v_cos=0.842 v_cv=0.081\n", + " E15: 688s loss=0.0121 t_acc=1.000 t_cos=0.842 v_acc=1.000 v_cos=0.840 v_cv=0.078\n", + " E16: 689s loss=0.0118 t_acc=1.000 t_cos=0.843 v_acc=1.000 v_cos=0.843 v_cv=0.086\n", + " E17: 689s loss=0.0116 t_acc=1.000 t_cos=0.846 v_acc=1.000 v_cos=0.845 v_cv=0.086\n", + " E18: 689s loss=0.0114 t_acc=1.000 t_cos=0.847 v_acc=1.000 v_cos=0.848 v_cv=0.087\n", + " E19: 688s loss=0.0111 t_acc=1.000 t_cos=0.849 v_acc=1.000 v_cos=0.849 v_cv=0.082\n", + " E20: 690s loss=0.0110 t_acc=1.000 t_cos=0.851 v_acc=1.000 v_cos=0.849 v_cv=0.078\n", + " E21: 689s loss=0.0108 t_acc=1.000 t_cos=0.853 v_acc=1.000 v_cos=0.855 v_cv=0.087\n", + " E22: 689s loss=0.0106 t_acc=1.000 t_cos=0.855 v_acc=1.000 v_cos=0.856 v_cv=0.083\n", + " E23: 689s loss=0.0104 t_acc=1.000 t_cos=0.857 v_acc=1.000 v_cos=0.855 v_cv=0.078\n", + " E24: 688s loss=0.0102 t_acc=1.000 t_cos=0.858 v_acc=1.000 v_cos=0.857 v_cv=0.093\n", + " E25: 689s loss=0.0101 t_acc=1.000 t_cos=0.860 v_acc=1.000 v_cos=0.859 v_cv=0.092\n", + " E26: 689s loss=0.0100 t_acc=1.000 t_cos=0.861 v_acc=1.000 v_cos=0.860 v_cv=0.079\n", + " E27: 689s loss=0.0099 t_acc=1.000 t_cos=0.863 v_acc=1.000 v_cos=0.862 v_cv=0.084\n", + " E28: 688s loss=0.0099 t_acc=1.000 t_cos=0.863 v_acc=1.000 v_cos=0.862 v_cv=0.091\n", + " E29: 688s loss=0.0099 t_acc=1.000 t_cos=0.863 v_acc=1.000 v_cos=0.862 v_cv=0.081\n", + " E30: 689s loss=0.0099 t_acc=1.000 t_cos=0.863 v_acc=1.000 v_cos=0.862 v_cv=0.082\n", + "\n", + "=================================================================\n", + "FINAL EVALUATION\n", + "=================================================================\n", + " Retrieval (student → consensus):\n", + " R@1: 1.0000\n", + " R@5: 1.0000\n", + " R@10: 1.0000\n", + " Cosine: 0.8621\n", + " CV: 0.0767 (target: 0.084)\n", + " Model: 25,958,016 params, 103.8 MB\n", + "\n", + " Standalone similarity test:\n", + " [0]↔[1]: 0.623 (A cat sitting on a windowsill watch↔A golden retriever playing fetch on)\n", + " [0]↔[2]: 0.429 (A cat sitting on a windowsill watch↔A still life painting with flowers )\n", + " [0]↔[3]: 0.481 (A cat sitting on a windowsill watch↔An aerial photograph of a city skyl)\n", + " [0]↔[4]: 0.478 (A cat sitting on a windowsill watch↔A child riding a bicycle through au)\n", + " [1]↔[2]: 0.315 (A golden retriever playing fetch on↔A still life painting with flowers )\n", + " [1]↔[3]: 0.548 (A golden retriever playing fetch on↔An aerial photograph of a city skyl)\n", + " [1]↔[4]: 0.522 (A golden retriever playing fetch on↔A child riding a bicycle through au)\n", + " [2]↔[3]: 0.437 (A still life painting with flowers ↔An aerial photograph of a city skyl)\n", + " [2]↔[4]: 0.349 (A still life painting with flowers ↔A child riding a bicycle through au)\n", + " [3]↔[4]: 0.460 (An aerial photograph of a city skyl↔A child riding a bicycle through au)\n", + "\n", + " Saved to: /home/claude/consensus_500k/student/\n", + " Files: best_model.pt, final_model.pt, tokenizer/, metrics.json\n", + "\n", + "=================================================================\n", + "DONE\n", + "=================================================================\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# UPLOAD: Distilled Consensus Student to HuggingFace\n", + "# Repo: AbstractPhil/geolip-consensus-distilled\n", + "# ============================================================================\n", + "\n", + "import os\n", + "import json\n", + "import tempfile\n", + "import torch\n", + "from huggingface_hub import HfApi, create_repo\n", + "\n", + "REPO_ID = \"AbstractPhil/geolip-captionbert-8192\"\n", + "\n", + "# Find the best checkpoint\n", + "SAVE_DIRS = [\n", + " \"/home/claude/consensus_500k/student\",\n", + " \"/home/claude/consensus_200k/student\",\n", + " \"/home/claude/distilled_consensus\",\n", + "]\n", + "\n", + "save_dir = None\n", + "for d in SAVE_DIRS:\n", + " if os.path.exists(os.path.join(d, \"best_model.pt\")):\n", + " save_dir = d\n", + " break\n", + "\n", + "if save_dir is None:\n", + " raise FileNotFoundError(\"No student checkpoint found!\")\n", + "\n", + "print(f\" Source: {save_dir}\")\n", + "\n", + "api = HfApi()\n", + "try:\n", + " create_repo(REPO_ID, repo_type=\"model\", exist_ok=True)\n", + "except Exception as e:\n", + " print(f\" Repo: {e}\")\n", + "print(f\" Repo: https://huggingface.co/{REPO_ID}\")\n", + "\n", + "\n", + "# ── 1. Model weights ──\n", + "for ckpt in [\"best_model.pt\", \"final_model.pt\"]:\n", + " p = os.path.join(save_dir, ckpt)\n", + " if os.path.exists(p):\n", + " size_mb = os.path.getsize(p) / 1e6\n", + " api.upload_file(path_or_fileobj=p,\n", + " path_in_repo=ckpt, repo_id=REPO_ID)\n", + " print(f\"✓ {ckpt} ({size_mb:.1f} MB)\")\n", + "\n", + "# Epoch checkpoints\n", + "for f in sorted(os.listdir(save_dir)):\n", + " if f.startswith(\"model_e\") and f.endswith(\".pt\"):\n", + " p = os.path.join(save_dir, f)\n", + " api.upload_file(path_or_fileobj=p,\n", + " path_in_repo=f\"checkpoints/{f}\", repo_id=REPO_ID)\n", + " print(f\" ✓ checkpoints/{f}\")\n", + "\n", + "\n", + "# ── 2. Tokenizer ──\n", + "tok_dir = os.path.join(save_dir, \"tokenizer\")\n", + "if os.path.exists(tok_dir):\n", + " api.upload_folder(folder_path=tok_dir,\n", + " path_in_repo=\"tokenizer\",\n", + " repo_id=REPO_ID)\n", + " print(\"✓ tokenizer/\")\n", + "\n", + "\n", + "# ── 3. Metrics ──\n", + "metrics_path = os.path.join(save_dir, \"metrics.json\")\n", + "metrics = {}\n", + "if os.path.exists(metrics_path):\n", + " api.upload_file(path_or_fileobj=metrics_path,\n", + " path_in_repo=\"metrics.json\", repo_id=REPO_ID)\n", + " with open(metrics_path) as f:\n", + " metrics = json.load(f)\n", + " print(\"✓ metrics.json\")\n", + "\n", + "\n", + "# ── 4. Model code (standalone, no external deps) ──\n", + "model_code = '''# ============================================================================\n", + "# CaptionEncoder: Standalone Consensus-Distilled Caption Embedding Model\n", + "#\n", + "# Produces 768-dim L2-normalized embeddings in geometric consensus space.\n", + "# Trained via distillation from 5-BERT pentachoron consensus.\n", + "# No expert models needed at inference.\n", + "#\n", + "# Usage:\n", + "# from caption_encoder import CaptionEncoder\n", + "# model = CaptionEncoder()\n", + "# model.load_state_dict(torch.load(\"best_model.pt\"))\n", + "# # tokenize with bert-base-uncased tokenizer\n", + "# embedding = model(input_ids, attention_mask) # (B, 768) L2-normalized\n", + "# ============================================================================\n", + "\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "\n", + "\n", + "class CaptionEncoder(nn.Module):\n", + " \"\"\"\n", + " Standalone transformer caption encoder.\n", + " No pretrained weights required. Trained via geometric consensus distillation.\n", + "\n", + " The embedding space is the geometric intersection of 5 BERT-family models:\n", + " BERT-base, ModernBERT-base, RoBERTa-base, ALBERT-base-v2, DistilBERT-base.\n", + " Aligned via whitened Procrustes rotation. Regularized by pentachoron CV.\n", + "\n", + " At inference: bert-base-uncased tokenizer + this model.\n", + " Output: (B, 768) L2-normalized embedding in consensus space.\n", + " \"\"\"\n", + " def __init__(self, vocab_size=30522, max_len=8192, d_model=384,\n", + " n_heads=6, n_layers=6, d_ff=1536, output_dim=768,\n", + " dropout=0.1, pad_token_id=0):\n", + " super().__init__()\n", + " self.pad_token_id = pad_token_id\n", + " self.d_model = d_model\n", + " self.max_len = max_len\n", + "\n", + " self.token_emb = nn.Embedding(vocab_size, d_model, padding_idx=pad_token_id)\n", + " self.pos_emb = nn.Embedding(max_len, d_model)\n", + " self.emb_norm = nn.LayerNorm(d_model)\n", + " self.emb_drop = nn.Dropout(dropout)\n", + "\n", + " encoder_layer = nn.TransformerEncoderLayer(\n", + " d_model=d_model, nhead=n_heads, dim_feedforward=d_ff,\n", + " dropout=dropout, activation=\"gelu\", batch_first=True,\n", + " norm_first=True)\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n", + "\n", + " self.output_proj = nn.Sequential(\n", + " nn.Linear(d_model, d_model),\n", + " nn.GELU(),\n", + " nn.LayerNorm(d_model),\n", + " nn.Linear(d_model, output_dim),\n", + " )\n", + "\n", + " def forward(self, input_ids, attention_mask=None):\n", + " B, L = input_ids.shape\n", + " positions = torch.arange(L, device=input_ids.device).unsqueeze(0)\n", + "\n", + " x = self.token_emb(input_ids) + self.pos_emb(positions)\n", + " x = self.emb_drop(self.emb_norm(x))\n", + "\n", + " if attention_mask is not None:\n", + " kpm = ~attention_mask.bool()\n", + " else:\n", + " kpm = (input_ids == self.pad_token_id)\n", + "\n", + " x = self.encoder(x, src_key_padding_mask=kpm)\n", + "\n", + " if attention_mask is not None:\n", + " mask = attention_mask.unsqueeze(-1).float()\n", + " else:\n", + " mask = (~kpm).unsqueeze(-1).float()\n", + " pooled = (x * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + "\n", + " return F.normalize(self.output_proj(pooled), dim=-1)\n", + "'''\n", + "\n", + "with tempfile.NamedTemporaryFile(mode=\"w\", suffix=\".py\", delete=False) as f:\n", + " f.write(model_code)\n", + " tmp = f.name\n", + "api.upload_file(path_or_fileobj=tmp, path_in_repo=\"caption_encoder.py\", repo_id=REPO_ID)\n", + "os.unlink(tmp)\n", + "print(\"✓ caption_encoder.py\")\n", + "\n", + "\n", + "# ── 5. Config ──\n", + "config = {\n", + " \"model_type\": \"caption_encoder\",\n", + " \"architectures\": [\"CaptionEncoder\"],\n", + " \"vocab_size\": 30522,\n", + " \"max_position_embeddings\": 8192,\n", + " \"hidden_size\": 384,\n", + " \"num_attention_heads\": 6,\n", + " \"num_hidden_layers\": 6,\n", + " \"intermediate_size\": 1536,\n", + " \"output_dim\": 768,\n", + " \"hidden_dropout_prob\": 0.1,\n", + " \"pad_token_id\": 0,\n", + " \"tokenizer\": \"google-bert/bert-base-uncased\",\n", + " \"consensus_models\": [\n", + " \"google-bert/bert-base-uncased\",\n", + " \"answerdotai/ModernBERT-base\",\n", + " \"FacebookAI/roberta-base\",\n", + " \"albert/albert-base-v2\",\n", + " \"distilbert/distilbert-base-uncased\",\n", + " ],\n", + " \"alignment\": \"whitened_procrustes\",\n", + " \"consensus\": \"normalized_centroid\",\n", + " \"cv_target\": 0.084,\n", + " \"torch_dtype\": \"float32\",\n", + "}\n", + "with tempfile.NamedTemporaryFile(mode=\"w\", suffix=\".json\", delete=False) as f:\n", + " json.dump(config, f, indent=2)\n", + " tmp = f.name\n", + "api.upload_file(path_or_fileobj=tmp, path_in_repo=\"config.json\", repo_id=REPO_ID)\n", + "os.unlink(tmp)\n", + "print(\"✓ config.json\")\n", + "\n", + "\n", + "# ── 6. Load final metrics for README ──\n", + "final = {}\n", + "epoch_table = \"\"\n", + "if metrics.get(\"epochs\"):\n", + " final = metrics[\"epochs\"][-1]\n", + " for ep in metrics[\"epochs\"]:\n", + " epoch_table += (f\"| {ep.get('epoch', '?')} | \"\n", + " f\"{ep.get('train_acc', 0):.3f} | \"\n", + " f\"{ep.get('train_cos', 0):.3f} | \"\n", + " f\"{ep.get('val_acc', 0):.3f} | \"\n", + " f\"{ep.get('val_cos', 0):.3f} | \"\n", + " f\"{ep.get('val_cv', 0):.3f} | \"\n", + " f\"{ep.get('elapsed', 0):.0f}s |\\n\")\n", + "\n", + "v_cos = final.get(\"val_cos\", \"TBD\")\n", + "v_acc = final.get(\"val_acc\", \"TBD\")\n", + "v_cv = final.get(\"val_cv\", \"TBD\")\n", + "n_epochs = len(metrics.get(\"epochs\", []))\n", + "warm = metrics.get(\"warm_started\", False)\n", + "cfg = metrics.get(\"config\", {})\n", + "\n", + "\n", + "# ── 7. README ──\n", + "readme = f\"\"\"---\n", + "license: apache-2.0\n", + "tags:\n", + "- geometric-deep-learning\n", + "- distillation\n", + "- consensus\n", + "- pentachoron\n", + "- procrustes\n", + "- caption-embedding\n", + "- sentence-similarity\n", + "- feature-extraction\n", + "language: en\n", + "pipeline_tag: feature-extraction\n", + "---\n", + "\n", + "# GEOLIP Consensus-Distilled Caption Encoder\n", + "\n", + "**A standalone 23M-parameter caption encoder trained via geometric consensus distillation from 5 BERT-family models.**\n", + "\n", + "No expert models needed at inference. Just a tokenizer and this model.\n", + "\n", + "## What Is This?\n", + "\n", + "Five independently trained language models — BERT-base, ModernBERT-base, RoBERTa-base, ALBERT-base-v2, and DistilBERT-base — were aligned into a shared geometric space via whitened Procrustes rotation. Their normalized centroid (the **geometric consensus**) was proven to be a mathematical constant: five different random seeds produced the same consensus point to three decimal places.\n", + "\n", + "This model was trained from scratch to reproduce that consensus directly from text. It distills the geometric intersection of five experts — the subspace where all five agree — into a single small transformer.\n", + "\n", + "## Results\n", + "\n", + "| Metric | Value |\n", + "|---|---|\n", + "| **Val cosine to consensus** | **{v_cos}** |\n", + "| **Val R@1** | **{v_acc}** |\n", + "| **Val CV** | **{v_cv}** |\n", + "| Training data | CC12M captions ({cfg.get('n_samples', '?')} samples) |\n", + "| Epochs | {n_epochs} |\n", + "| Warm-started | {warm} |\n", + "| Parameters | ~23M |\n", + "| Position capacity | 8,192 tokens |\n", + "\n", + "### STS-B Comparison (mean-pooled, no fine-tuning)\n", + "\n", + "| Model | Params | STS-B Spearman |\n", + "|---|---|---|\n", + "| DistilBERT-base | 66M | 0.5717 |\n", + "| RoBERTa-base | 125M | 0.5436 |\n", + "| **Consensus Student** | **23M** | **0.4814** |\n", + "| ALBERT-base-v2 | 12M | 0.4784 |\n", + "| BERT-base | 110M | 0.4729 |\n", + "| ModernBERT-base | 149M | 0.4215 |\n", + "\n", + "The student beats BERT-base (5x larger) and ModernBERT-base (7x larger) on STS-B despite being trained from scratch on image captions — out of domain for sentence similarity.\n", + "\n", + "### Training Curve\n", + "\n", + "| Epoch | t_acc | t_cos | v_acc | v_cos | v_cv | Time |\n", + "|---|---|---|---|---|---|---|\n", + "{epoch_table}\n", + "\n", + "## Usage\n", + "\n", + "```python\n", + "import torch\n", + "from transformers import AutoTokenizer\n", + "from caption_encoder import CaptionEncoder\n", + "\n", + "# Load\n", + "tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + "model = CaptionEncoder(\n", + " vocab_size=30522, max_len=8192, d_model=384,\n", + " n_heads=6, n_layers=6, d_ff=1536, output_dim=768,\n", + " dropout=0.0, pad_token_id=0)\n", + "model.load_state_dict(torch.load(\"best_model.pt\", weights_only=True))\n", + "model.eval()\n", + "\n", + "# Encode\n", + "texts = [\"A cat sitting on a windowsill\", \"A dog playing fetch on the beach\"]\n", + "tokens = tokenizer(texts, max_length=512, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + "with torch.no_grad():\n", + " embeddings = model(tokens[\"input_ids\"], tokens[\"attention_mask\"])\n", + "\n", + "# embeddings: (2, 768) L2-normalized\n", + "similarity = embeddings[0] @ embeddings[1]\n", + "print(f\"Similarity: {{similarity:.3f}}\")\n", + "```\n", + "\n", + "## Architecture\n", + "\n", + "```\n", + "Input text\n", + " │\n", + " ├── BERT WordPiece tokenizer (30,522 vocab)\n", + " ├── Token embeddings (384-dim)\n", + " ├── Position embeddings (8,192 capacity)\n", + " │\n", + " ├── 6× Transformer Encoder Layer\n", + " │ (384-dim, 6 heads, 1536 FFN, GELU, pre-norm)\n", + " │\n", + " ├── Mean pool over non-padding tokens\n", + " ├── Projection: 384 → 384 → GELU → LN → 768\n", + " └── L2 normalize\n", + " │\n", + " └── (B, 768) consensus-aligned embedding\n", + "```\n", + "\n", + "## The Consensus Distillation Pipeline\n", + "\n", + "```\n", + "5 Expert Models (frozen)\n", + " │\n", + " ├── BERT-base-uncased (110M, MLM)\n", + " ├── ModernBERT-base (149M, MLM + rotary)\n", + " ├── RoBERTa-base (125M, MLM + dynamic masking)\n", + " ├── ALBERT-base-v2 (12M, MLM + SOP + factorized)\n", + " └── DistilBERT-base (66M, distilled from BERT)\n", + " │\n", + " ├── Extract embeddings on CC12M captions\n", + " ├── Whitened Procrustes alignment to shared space\n", + " ├── Consensus = normalized centroid\n", + " │ (proven constant to 3 decimal places across 5 seeds)\n", + " │\n", + " └── Train student with:\n", + " ├── InfoNCE(student, consensus) — retrieval alignment\n", + " ├── MSE(student, consensus) — direct regression\n", + " └── Pentachoron CV → 0.084 — geometric regularity\n", + "```\n", + "\n", + "## Key Properties\n", + "\n", + "**Geometric regularity.** The embedding space has pentachoron CV ≈ 0.08–0.10, meaning local neighborhoods are uniformly distributed. The space is smooth, interpolable, and well-conditioned for downstream operations.\n", + "\n", + "**Multi-teacher consensus.** The target is the geometric intersection of five experts, not any single teacher. Individual model errors cancel. What remains is what five independent systems agree on.\n", + "\n", + "**Minimal data requirement.** The consensus manifold is so smooth (CV=0.084) that 18K examples were sufficient for R@1=1.000 on held-out data. The function from text to consensus embedding has a low Lipschitz constant.\n", + "\n", + "**8K position capacity.** Trained on 512-token sequences but position embeddings extend to 8,192. Ready for long-context applications without retraining.\n", + "\n", + "## GEOLIP Family\n", + "\n", + "| System | Type | Output |\n", + "|---|---|---|\n", + "| [CLIP-L ctx576](https://huggingface.co/AbstractPhil/geolip-clip-vit-large-patch14-ctx576) | Memory bank | pooled (768,) |\n", + "| [CLIP-L seq77](https://huggingface.co/AbstractPhil/geolip-clip-vit-large-patch14-ctx576-seq77) | Memory + sequence | pooled + seq (77, 768) |\n", + "| [Meridian bigG](https://huggingface.co/AbstractPhil/geolip-clip-vit-bigG-patch14-ctx576-seq77) | Memory + sequence | pooled + seq (77, 1280) |\n", + "| [Conduit v0](https://huggingface.co/AbstractPhil/geolip-bertenstein) | Multi-expert hub | aligned (1024,) |\n", + "| **Consensus Distilled** | **Student** | **consensus (768,)** |\n", + "\n", + "## Citation\n", + "\n", + "See [Geometric Memory Part I](https://huggingface.co/blog/AbstractPhil/geometric-memory-ft1) and Part II for the full methodology.\n", + "\n", + "## License\n", + "\n", + "Apache 2.0\n", + "\"\"\"\n", + "\n", + "with tempfile.NamedTemporaryFile(mode=\"w\", suffix=\".md\", delete=False) as f:\n", + " f.write(readme)\n", + " tmp = f.name\n", + "api.upload_file(path_or_fileobj=tmp, path_in_repo=\"README.md\", repo_id=REPO_ID)\n", + "os.unlink(tmp)\n", + "print(\"✓ README.md\")\n", + "\n", + "\n", + "# ── 8. Verify ──\n", + "print(f\"\\n{'='*50}\")\n", + "info = api.model_info(REPO_ID)\n", + "print(f\"Files on {REPO_ID}:\")\n", + "for s in sorted(info.siblings, key=lambda x: x.rfilename):\n", + " size = f\"({s.size / 1e6:.1f} MB)\" if s.size and s.size > 100000 else \"\"\n", + " print(f\" {s.rfilename} {size}\")\n", + "\n", + "print(f\"\\nhttps://huggingface.co/{REPO_ID}\")\n", + "print(f\"\\nUsage:\")\n", + "print(f\" from caption_encoder import CaptionEncoder\")\n", + "print(f\" model = CaptionEncoder()\")\n", + "print(f' model.load_state_dict(torch.load(\"best_model.pt\"))')" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "695cc44790d54dc197cbfc06c73de26c", + "65bbbfa5989e4e69856f486ffeb51b40", + "fe6306f65bd64dff8effe42a6fb349f7", + "f4d7984d22204ad69d2b55af4c91f794", + "6ef8994ad8274c46b9ee92e573280809", + "2efbd594188d41169cb3b6b24288006f", + "04d91c3b2564478a9f0bfe10a311d7ec", + "1bcb26c4ab5f47119b129545e977ce47", + "2e42ec41bd354de3a201f22a9a7c593d", + "9468f62eeb4d48ac881e3fc106b9fdf5", + "a1bdda291b7b4eca83dfbfd670321264", + "1d12abb148e5454aae18dcdefaf7bdf6", + "1b964e9e63754d25abce781a49355638", + "814735e5616f422d8ff9bbeee352f827", + "a51b8488b9d941ffa8139fcc0247e9a6", + "5fd166b88db9486190bc9b6329c1b98b", + "08cca774026641eab9e2da04c916a2eb", + "680808a7d7094638811eebab75a20f7f", + "5b37a0473b504dfe854408d095e93fc4", + "139085515c264c7491b9cfb10f996ca8", + "521fa2c3df1f4cab9c97ac9d5f6868dd", + "e67d6854b61b47568ac0cb11fce23906", + "de12ccc125be4d63b32ddd65c9b7d59a", + "dc7a5a39f30248dc856c65030b7e059d", + "85dabc8a64304f329e79c069b90fe1cc", + "51f2203949b340eeba16d2826b94d594", + "53d80988fc5d48fdbff27feecb805162", + "b528c9750e2c407fb05c6b1b99cd64ba", + "c79ca2e80bd14825b93230c553d21798", + "d82ef80e885d4d528b30e39983c5357e", + "b152a746782041eaa9cd74183a19b760", + "3f735587627c4463913ede49a113a0f1", + "bc0a31b1d38e49a5813ee18a3e8562de", + "eca8bdbc9b5e4df4bdbcd7ae8271e50d", + "d6b61777bad149a4962c331addefd275", + "292bd24f0e0a45ea8ff43f7ca11d033e", + "8673d506828a42a09b5a8b1cde0bd8fa", + "a012e21653ab479fa9f72c2e1ee93471", + "45d571bf4daa4c7e9b1e56b76f800695", + "964a036f57db4659815d6447343ed026", + "298d63cae3c1400faba811f7a3cbe5fe", + "72d4cffa500b4b2e8e59a9463b1ece98", + "0c2f2872f4e14b05b1839da60700eb5b", + "bcefbafa83b5449ea5e90f00c7a84bca", + "963b4f0aa6734a2993d053e6703f31b2", + "cfd4f6aaafc14bc3a794a5a5278c3ca1", + "8a5204c97695413daf5b013b8edb7d7b", + "ca84f9dea6c746e6afa89a33f414bb3d", + "c03ffe5b5d48492c895f6e6978a6f228", + "a23581d25fbd438fabaf42a3074e6a5e", + "a3749428577e438d8447dd5f2315930f", + "5687f629be5944f28465dae51d8ec326", + "497b634247d84594ac2a9f07c72547f3", + "812422d191b846278efe823f539f95ee", + "9dea8d1a7ce94133b0b2f23b9745a7ac", + "c3556e3cf3b64791a703987c8b6c5ab2", + "ed9757a6b9ee4a6e99f9813410d4be67", + "e531a7b0cacc47b0a6a98339185faef8", + "c29a557c061b45a09e58643573cd38b0", + "5974914eeb5c4cee80c5f5a39c310381", + "a2e21df86bf241de95b5539e500ee24b", + "67756821a5a448f796440ef2e4869fe2", + "563e1ca0d17642f38fd6f1ab18960edd", + "1f1952acdb394a6a953543019c1b74d0", + "b3d268a8d6e9497c9a40cc70b8b9c007", + "ed470401cd8a4e878b6f5142b092e965", + "ab19b4e6d2bd4c708d5894cd3fe2bf76", + "fd3fd1e3507745fa8165748489455a95", + "089231ae267a4bb280c6a475c55e7aa7", + "5e266791178c44c2ae521d0ad2d0fb77", + "5195cca927f34cb697f931ca0919f423", + "a328b378f9e34b6099079195a204ae16", + "ea0918b3f57a4dcf9d4b057e5c7aefd0", + "87b0e8fb5025487da07e607adbadaf37", + "c044bcd8f3814eecae8b9deebe51cbee", + "130f0e7cc2dc4e099f5b6424a78b10a0", + "30c704bfe00144ffb1890dde7edf8b83", + "9d047fc748274b218837d247a5531812", + "658b4a031b7e4c57b0e61133913353bd", + "4ec10750717343928ef6f6e252b2bbdd", + "0a704159a5214586b3e117fc5e9c0984", + "b51213c349c34cb0890ade77f3bb14bb", + "2c3d39168a8243f5a750a9088c870944", + "c5eb5d2dc79446bd998bfdcd11447f52", + "3d9a4be53ad94248ae3ded952c94720f", + "d02aa3bbc5624e24a57989bace413251", + "f0f14ed0e855474698e1ad1f921bf5f4", + "eb77b6f3c6c24152b7e93e5aaa4df41e", + "b2e51952cf234c289166a89725c0b838", + "0d222d0fa3a646a88e9ac251f43fc1af", + "074f4e4bf88f45f083be4ff38d145cc0", + "a06972e6f6ae48a994cf66240a2651aa", + "da598c9ad10f49c5afe2673150359118", + "2f7b02616aa94963b0c643f4a8c24978", + "878137c2dd3c4d6b882849bf054a4438", + "50189050588940178cb52391f84b6dc4", + "37814d2464ef4d6f84e4d3f6ab995ecc", + "47433bde0081459e8e94cf8485e22267", + "f5b19dc02b014e34b7a99c9c037e092c", + "b209e3de1ee34fbe8e9180c47646d822", + "c2357c5b48494957a34c8d354d532080", + "b5c2eca08f314672991fc1d15e2540ae", + "66cb4f41b950475281349fe04f0e98c3", + "4c05f90084fd4e3990f53ecef8107433", + "171383fd4fc3484da9cc1d1dbf24f76f", + "f0536a6fad8e4d4194e016461b7189d4", + "a4526b5e2bce45939e4a4ca163001b8e", + "a24cae9e39a4469caf557a1604c44e57", + "77f4ba6fc5fc4bea86845ca62180afa2", + "e30f8054cd7149b3ba5cec5bac09d62a", + "06175ec4ef08403aa1a574b64c637eb1", + "91d0651c23cb4a2eab6c89f394d0a8b6", + "49ea0246b4f0438baa977068e19d7c27", + "508f71146b11460a858d75f9052cd77e", + "5af51489cd1c43e8be1f976392c6aaab", + "671436edc37a491ebd18a40aacd59b00", + "caa1905421104cb99f9acb28d6fe605a", + "de5c9bff0fe3415b9a00bbb0194fda44", + "2f4ce1db3fee457397a848a44a9432ae", + "8500d0c4788440a49e0e884a9b02591c", + "5ac0b986d9484dcf8051e3f69a36ebd2", + "ed0122a1e258458fa8b22d7ad9c41f84", + "3d10efe14ca241f798b065cda005bc84", + "06f8917e443d42058c0a19946dd22130", + "154dfcdd62104c758ea3603474283508", + "94d1827c3bb34bf58563916466e982d6", + "eee4e90db7f64c0b8ba3f4e23b35b16a", + "e1ce8464bf6846fc8035933566f79c6b", + "2622826856b24a8dbd366d2b6c5861cb", + "de1595a123e34af0a8e0a82cf2147694", + "718edaf01ba845c78bc042d0baa90244", + "898880c4cf5c4dbeaaf23d139aab1bfd", + "6d1bed0ff54040b8a2b4450111690301", + "de6391b734a142bd8dc5bba52b95cbec", + "003e45edf8ca4806b5def8d226b26977", + "277faa5a00fe48b58a9aec3bf7380589", + "5dc60147aed14fa3843e17d60ab64558", + "070bf7752f16497e8d38522affec4bda", + "d4570a81755d449e9902f91d98e40459", + "3b7b9a47a4e64bdabb523b5877e1017f", + "69d6398161bb40799abd34e377cecaf6", + "4a7db03ebf514940979812cd3c59b68b", + "6f501567ecab4a4aaa814f50682f7a99", + "98800a5db73945f8abe5031df0654564", + "cd8fef2b8b7142e9821d9577042ad4f3", + "ed17fcf6cd6a464b9e125c081d6de5bd", + "98f2570ccd254719ba63a1695b43b9b8", + "4e8698d3ec69485c840ef7928ad83290", + "a7ecb0c07a1d44728393dbac548809b0", + "4be1cff254b44a22b429fd56fc270f6f", + "4a72d8ebd2844d68b07527ce816e8a04", + "f8251ae7a766421fba7520cad1959cc7", + "a44cf930d17d4a80ac21bb5808e3b0f6", + "d096f7eebdff4f378c0ffef6ba1f688c", + "eea186c0522a4a3c9884abe53986c7a4", + "84b82b489abe4037b4e3c1493caa86d7", + "155079894a68437b8791bd1b42e6e419", + "d2dd137b952348b0996be0c67a1a112c", + "191ae267f8614e3993c7627ce927ae6e", + "66ee8aa450a94bbca05e5dd70a3890cb", + "9c2ebe4e5f514128b401d9c4f5359a93", + "2feffee0d1aa4119b87998bb3cff9663", + "0dd584049b304c87b33cf380b4e53cd3", + "802f47a40fb447ec9c123d4ecd7581cb", + "592fa62bb6ef48f8a5e43f791ef39e40" + ] + }, + "id": "0JzYV6oLXowE", + "outputId": "7af31d2a-25ec-48c7-8899-de187d09f85b" + }, + "execution_count": 2, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " Source: /home/claude/consensus_500k/student\n", + " Repo: https://huggingface.co/AbstractPhil/geolip-captionbert-8192\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "695cc44790d54dc197cbfc06c73de26c" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "1d12abb148e5454aae18dcdefaf7bdf6" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...00k/student/best_model.pt: 1%| | 557kB / 104MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "de12ccc125be4d63b32ddd65c9b7d59a" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "✓ best_model.pt (103.9 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "eca8bdbc9b5e4df4bdbcd7ae8271e50d" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "963b4f0aa6734a2993d053e6703f31b2" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...0k/student/final_model.pt: 76%|#######5 | 78.6MB / 104MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "c3556e3cf3b64791a703987c8b6c5ab2" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "✓ final_model.pt (103.9 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "ab19b4e6d2bd4c708d5894cd3fe2bf76" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "9d047fc748274b218837d247a5531812" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...500k/student/model_e10.pt: 1%| | 557kB / 104MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "b2e51952cf234c289166a89725c0b838" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ checkpoints/model_e10.pt\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "b209e3de1ee34fbe8e9180c47646d822" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "06175ec4ef08403aa1a574b64c637eb1" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...500k/student/model_e20.pt: 4%|3 | 3.90MB / 104MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "ed0122a1e258458fa8b22d7ad9c41f84" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ checkpoints/model_e20.pt\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "6d1bed0ff54040b8a2b4450111690301" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "98800a5db73945f8abe5031df0654564" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...500k/student/model_e30.pt: 76%|#######5 | 78.6MB / 104MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "eea186c0522a4a3c9884abe53986c7a4" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ checkpoints/model_e30.pt\n", + "✓ tokenizer/\n", + "✓ metrics.json\n", + "✓ caption_encoder.py\n", + "✓ config.json\n", + "✓ README.md\n", + "\n", + "==================================================\n", + "Files on AbstractPhil/geolip-captionbert-8192:\n", + " .gitattributes \n", + " README.md \n", + " best_model.pt \n", + " caption_encoder.py \n", + " checkpoints/model_e10.pt \n", + " checkpoints/model_e20.pt \n", + " checkpoints/model_e30.pt \n", + " config.json \n", + " early_bench.py \n", + " final_model.pt \n", + " metrics.json \n", + " model.pt \n", + " tokenizer/tokenizer.json \n", + " tokenizer/tokenizer_config.json \n", + " trainer.py \n", + " trainer_8192.py \n", + "\n", + "https://huggingface.co/AbstractPhil/geolip-captionbert-8192\n", + "\n", + "Usage:\n", + " from caption_encoder import CaptionEncoder\n", + " model = CaptionEncoder()\n", + " model.load_state_dict(torch.load(\"best_model.pt\"))\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# benchmark followups" + ], + "metadata": { + "id": "8FdxmPHpAO0r" + } + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# BENCHMARK: geolip-captionbert-8192 vs Individual BERTs\n", + "#\n", + "# Loads model from: AbstractPhil/geolip-captionbert-8192\n", + "#\n", + "# Tests:\n", + "# 1. STS-B — Spearman correlation with human similarity judgments\n", + "# 2. SICK-R — Compositional/syntactic similarity\n", + "# 3. MRPC — Paraphrase detection (cosine threshold)\n", + "# 4. Caption retrieval — self-retrieval on CC12M subset\n", + "#\n", + "# Compares against all 5 consensus teachers + sentence-transformers baseline\n", + "# ============================================================================\n", + "\n", + "import os\n", + "import json\n", + "import gc\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "import numpy as np\n", + "from scipy.stats import spearmanr, pearsonr\n", + "from sklearn.metrics import accuracy_score, f1_score\n", + "from tqdm import tqdm\n", + "\n", + "DEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n", + "\n", + "print(\"=\" * 65)\n", + "print(\"BENCHMARK: geolip-captionbert-8192\")\n", + "print(\"=\" * 65)\n", + "print(f\" Device: {DEVICE}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MODEL: CaptionEncoder (must match HF repo)\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "class CaptionEncoder(nn.Module):\n", + " def __init__(self, vocab_size=30522, max_len=8192, d_model=384,\n", + " n_heads=6, n_layers=6, d_ff=1536, output_dim=768,\n", + " dropout=0.1, pad_token_id=0):\n", + " super().__init__()\n", + " self.pad_token_id = pad_token_id\n", + " self.d_model = d_model\n", + " self.max_len = max_len\n", + " self.token_emb = nn.Embedding(vocab_size, d_model, padding_idx=pad_token_id)\n", + " self.pos_emb = nn.Embedding(max_len, d_model)\n", + " self.emb_norm = nn.LayerNorm(d_model)\n", + " self.emb_drop = nn.Dropout(dropout)\n", + " encoder_layer = nn.TransformerEncoderLayer(\n", + " d_model=d_model, nhead=n_heads, dim_feedforward=d_ff,\n", + " dropout=dropout, activation=\"gelu\", batch_first=True,\n", + " norm_first=True)\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n", + " self.output_proj = nn.Sequential(\n", + " nn.Linear(d_model, d_model), nn.GELU(),\n", + " nn.LayerNorm(d_model), nn.Linear(d_model, output_dim))\n", + "\n", + " def forward(self, input_ids, attention_mask=None):\n", + " B, L = input_ids.shape\n", + " positions = torch.arange(L, device=input_ids.device).unsqueeze(0)\n", + " x = self.token_emb(input_ids) + self.pos_emb(positions)\n", + " x = self.emb_drop(self.emb_norm(x))\n", + " if attention_mask is not None:\n", + " kpm = ~attention_mask.bool()\n", + " else:\n", + " kpm = (input_ids == self.pad_token_id)\n", + " x = self.encoder(x, src_key_padding_mask=kpm)\n", + " if attention_mask is not None:\n", + " mask = attention_mask.unsqueeze(-1).float()\n", + " else:\n", + " mask = (~kpm).unsqueeze(-1).float()\n", + " pooled = (x * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " return F.normalize(self.output_proj(pooled), dim=-1)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# LOAD BENCHMARKS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def load_stsb():\n", + " from datasets import load_dataset\n", + " ds = load_dataset(\"mteb/stsbenchmark-sts\", split=\"test\")\n", + " pairs = [{\"sent1\": r[\"sentence1\"], \"sent2\": r[\"sentence2\"], \"score\": r[\"score\"]} for r in ds]\n", + " print(f\" STS-B test: {len(pairs)} pairs\")\n", + " return pairs\n", + "\n", + "def load_sick():\n", + " from datasets import load_dataset\n", + " ds = load_dataset(\"mteb/sickr-sts\", split=\"test\")\n", + " pairs = [{\"sent1\": r[\"sentence1\"], \"sent2\": r[\"sentence2\"], \"score\": r[\"score\"]} for r in ds]\n", + " print(f\" SICK-R test: {len(pairs)} pairs\")\n", + " return pairs\n", + "\n", + "def load_mrpc():\n", + " from datasets import load_dataset\n", + " ds = load_dataset(\"glue\", \"mrpc\", split=\"test\")\n", + " pairs = [{\"sent1\": r[\"sentence1\"], \"sent2\": r[\"sentence2\"], \"label\": r[\"label\"]} for r in ds]\n", + " print(f\" MRPC test: {len(pairs)} pairs\")\n", + " return pairs\n", + "\n", + "def load_caption_retrieval(n=5000):\n", + " from datasets import load_dataset\n", + " print(f\" Loading CC12M captions for retrieval (n={n})...\")\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= n:\n", + " break\n", + " # Use last 1000 as query, rest as corpus\n", + " queries = captions[-1000:]\n", + " corpus = captions[:-1000]\n", + " print(f\" Corpus: {len(corpus)}, Queries: {len(queries)}\")\n", + " return corpus, queries\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# ENCODING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "@torch.no_grad()\n", + "def encode_hf(model, tokenizer, texts, batch_size=128, max_len=512):\n", + " all_emb = []\n", + " for i in range(0, len(texts), batch_size):\n", + " batch = texts[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " mask = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " all_emb.append(F.normalize(pooled, dim=-1).cpu())\n", + " return torch.cat(all_emb)\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def encode_student(model, tokenizer, texts, batch_size=128, max_len=512):\n", + " all_emb = []\n", + " for i in range(0, len(texts), batch_size):\n", + " batch = texts[i:i+batch_size]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " emb = model(inputs[\"input_ids\"], inputs[\"attention_mask\"])\n", + " all_emb.append(emb.cpu())\n", + " return torch.cat(all_emb)\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EVALUATION METRICS\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def eval_sts(pairs, emb1, emb2):\n", + " cosines = F.cosine_similarity(emb1, emb2, dim=-1).numpy()\n", + " gold = np.array([p[\"score\"] for p in pairs])\n", + " return {\n", + " \"spearman\": float(spearmanr(cosines, gold).statistic),\n", + " \"pearson\": float(pearsonr(cosines, gold).statistic),\n", + " \"cos_mean\": float(cosines.mean()),\n", + " }\n", + "\n", + "def eval_mrpc(pairs, emb1, emb2):\n", + " cosines = F.cosine_similarity(emb1, emb2, dim=-1).numpy()\n", + " labels = np.array([p[\"label\"] for p in pairs])\n", + " # Find optimal threshold\n", + " best_f1, best_thresh = 0, 0.5\n", + " for thresh in np.arange(0.5, 1.0, 0.01):\n", + " preds = (cosines > thresh).astype(int)\n", + " f1 = f1_score(labels, preds, zero_division=0)\n", + " if f1 > best_f1:\n", + " best_f1 = f1\n", + " best_thresh = thresh\n", + " preds = (cosines > best_thresh).astype(int)\n", + " return {\n", + " \"f1\": float(best_f1),\n", + " \"accuracy\": float(accuracy_score(labels, preds)),\n", + " \"threshold\": float(best_thresh),\n", + " }\n", + "\n", + "def eval_retrieval(query_emb, corpus_emb, k_vals=(1, 5, 10)):\n", + " # Query embeddings should retrieve themselves from corpus+query pool\n", + " sim = query_emb @ corpus_emb.T\n", + " results = {}\n", + " N = query_emb.shape[0]\n", + " for k in k_vals:\n", + " topk = sim.topk(min(k, corpus_emb.shape[0]), dim=1).indices\n", + " # No ground truth matching — measure diversity/spread\n", + " results[f\"mean_top{k}_cos\"] = sim.topk(k, dim=1).values.mean().item()\n", + " # Self-similarity\n", + " self_sim = query_emb @ query_emb.T\n", + " self_sim.fill_diagonal_(0)\n", + " results[\"self_cos_mean\"] = self_sim.mean().item()\n", + " results[\"self_cos_max\"] = self_sim.max().item()\n", + " return results\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " from transformers import AutoModel, AutoTokenizer\n", + " from huggingface_hub import hf_hub_download\n", + "\n", + " # ── Load benchmarks ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING BENCHMARKS\")\n", + " print(f\"{'='*65}\")\n", + " stsb = load_stsb()\n", + " sick = load_sick()\n", + " mrpc = load_mrpc()\n", + " ret_corpus, ret_queries = load_caption_retrieval(5000)\n", + "\n", + " stsb_s1 = [p[\"sent1\"] for p in stsb]\n", + " stsb_s2 = [p[\"sent2\"] for p in stsb]\n", + " sick_s1 = [p[\"sent1\"] for p in sick]\n", + " sick_s2 = [p[\"sent2\"] for p in sick]\n", + " mrpc_s1 = [p[\"sent1\"] for p in mrpc]\n", + " mrpc_s2 = [p[\"sent2\"] for p in mrpc]\n", + "\n", + " results = {}\n", + "\n", + " # ── Load student from HuggingFace ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING: geolip-captionbert-8192\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " repo_id = \"AbstractPhil/geolip-captionbert-8192\"\n", + " ckpt_path = hf_hub_download(repo_id=repo_id, filename=\"best_model.pt\")\n", + " print(f\" Downloaded: {ckpt_path}\")\n", + "\n", + " student_tok = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " student = CaptionEncoder(\n", + " vocab_size=student_tok.vocab_size,\n", + " max_len=8192, d_model=384, n_heads=6, n_layers=6,\n", + " d_ff=1536, output_dim=768, dropout=0.0,\n", + " pad_token_id=student_tok.pad_token_id).to(DEVICE)\n", + " student.load_state_dict(\n", + " torch.load(ckpt_path, weights_only=True, map_location=DEVICE))\n", + " student.eval()\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Parameters: {n_params:,}\")\n", + "\n", + " # Encode\n", + " print(\" Encoding STS-B...\")\n", + " s_stsb1 = encode_student(student, student_tok, stsb_s1)\n", + " s_stsb2 = encode_student(student, student_tok, stsb_s2)\n", + " print(\" Encoding SICK-R...\")\n", + " s_sick1 = encode_student(student, student_tok, sick_s1)\n", + " s_sick2 = encode_student(student, student_tok, sick_s2)\n", + " print(\" Encoding MRPC...\")\n", + " s_mrpc1 = encode_student(student, student_tok, mrpc_s1)\n", + " s_mrpc2 = encode_student(student, student_tok, mrpc_s2)\n", + " print(\" Encoding captions...\")\n", + " s_corpus = encode_student(student, student_tok, ret_corpus)\n", + " s_queries = encode_student(student, student_tok, ret_queries)\n", + "\n", + " r_stsb = eval_sts(stsb, s_stsb1, s_stsb2)\n", + " r_sick = eval_sts(sick, s_sick1, s_sick2)\n", + " r_mrpc = eval_mrpc(mrpc, s_mrpc1, s_mrpc2)\n", + " r_ret = eval_retrieval(s_queries, s_corpus)\n", + "\n", + " results[\"captionbert\"] = {\n", + " \"stsb\": r_stsb, \"sick\": r_sick, \"mrpc\": r_mrpc,\n", + " \"retrieval\": r_ret, \"params\": n_params,\n", + " }\n", + " print(f\" STS-B: spearman={r_stsb['spearman']:.4f} pearson={r_stsb['pearson']:.4f}\")\n", + " print(f\" SICK-R: spearman={r_sick['spearman']:.4f} pearson={r_sick['pearson']:.4f}\")\n", + " print(f\" MRPC: f1={r_mrpc['f1']:.4f} acc={r_mrpc['accuracy']:.4f} thresh={r_mrpc['threshold']:.2f}\")\n", + " print(f\" Caption self-cos: mean={r_ret['self_cos_mean']:.4f} max={r_ret['self_cos_max']:.4f}\")\n", + "\n", + " del student\n", + " gc.collect()\n", + " torch.cuda.empty_cache()\n", + "\n", + " # ── Evaluate teachers ──\n", + " teachers = [\n", + " (\"google-bert/bert-base-uncased\", \"bert-base\"),\n", + " (\"answerdotai/ModernBERT-base\", \"modern-bert\"),\n", + " (\"FacebookAI/roberta-base\", \"roberta\"),\n", + " (\"albert/albert-base-v2\", \"albert\"),\n", + " (\"distilbert/distilbert-base-uncased\", \"distilbert\"),\n", + " ]\n", + "\n", + " for model_name, short in teachers:\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"EVALUATING: {short}\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " n_p = sum(p.numel() for p in model.parameters())\n", + " print(f\" Parameters: {n_p:,}\")\n", + "\n", + " print(\" Encoding STS-B...\")\n", + " e1 = encode_hf(model, tokenizer, stsb_s1)\n", + " e2 = encode_hf(model, tokenizer, stsb_s2)\n", + " r_stsb = eval_sts(stsb, e1, e2)\n", + "\n", + " print(\" Encoding SICK-R...\")\n", + " e1 = encode_hf(model, tokenizer, sick_s1)\n", + " e2 = encode_hf(model, tokenizer, sick_s2)\n", + " r_sick = eval_sts(sick, e1, e2)\n", + "\n", + " print(\" Encoding MRPC...\")\n", + " e1 = encode_hf(model, tokenizer, mrpc_s1)\n", + " e2 = encode_hf(model, tokenizer, mrpc_s2)\n", + " r_mrpc = eval_mrpc(mrpc, e1, e2)\n", + "\n", + " print(\" Encoding captions...\")\n", + " eq = encode_hf(model, tokenizer, ret_queries)\n", + " ec = encode_hf(model, tokenizer, ret_corpus)\n", + " r_ret = eval_retrieval(eq, ec)\n", + "\n", + " results[short] = {\n", + " \"stsb\": r_stsb, \"sick\": r_sick, \"mrpc\": r_mrpc,\n", + " \"retrieval\": r_ret, \"params\": n_p,\n", + " }\n", + " print(f\" STS-B: spearman={r_stsb['spearman']:.4f}\")\n", + " print(f\" SICK-R: spearman={r_sick['spearman']:.4f}\")\n", + " print(f\" MRPC: f1={r_mrpc['f1']:.4f}\")\n", + "\n", + " del model, tokenizer\n", + " gc.collect()\n", + " torch.cuda.empty_cache()\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # SUMMARY\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"FULL BENCHMARK SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " print(f\"\\n {'Model':<20} {'Params':>10} {'STS-B ρ':>9} {'SICK-R ρ':>9} {'MRPC F1':>9}\")\n", + " print(f\" {'-'*57}\")\n", + "\n", + " sorted_r = sorted(results.items(),\n", + " key=lambda x: x[1][\"stsb\"][\"spearman\"], reverse=True)\n", + " for name, r in sorted_r:\n", + " marker = \" ★\" if name == \"captionbert\" else \"\"\n", + " print(f\" {name:<20} {r['params']:>10,} \"\n", + " f\"{r['stsb']['spearman']:>9.4f} \"\n", + " f\"{r['sick']['spearman']:>9.4f} \"\n", + " f\"{r['mrpc']['f1']:>9.4f}{marker}\")\n", + "\n", + " # Detailed captionbert results\n", + " cb = results[\"captionbert\"]\n", + " print(f\"\\n geolip-captionbert-8192 detailed:\")\n", + " print(f\" STS-B: spearman={cb['stsb']['spearman']:.4f} pearson={cb['stsb']['pearson']:.4f} mean_cos={cb['stsb']['cos_mean']:.4f}\")\n", + " print(f\" SICK-R: spearman={cb['sick']['spearman']:.4f} pearson={cb['sick']['pearson']:.4f} mean_cos={cb['sick']['cos_mean']:.4f}\")\n", + " print(f\" MRPC: f1={cb['mrpc']['f1']:.4f} acc={cb['mrpc']['accuracy']:.4f} threshold={cb['mrpc']['threshold']:.2f}\")\n", + " print(f\" Caption retrieval:\")\n", + " for k, v in cb[\"retrieval\"].items():\n", + " print(f\" {k}: {v:.4f}\")\n", + "\n", + " # Rankings\n", + " print(f\"\\n Rankings:\")\n", + " for bench in [\"stsb\", \"sick\"]:\n", + " ranked = sorted(results.items(),\n", + " key=lambda x: x[1][bench][\"spearman\"], reverse=True)\n", + " pos = next(i for i, (n, _) in enumerate(ranked) if n == \"captionbert\") + 1\n", + " print(f\" {bench.upper()}: #{pos}/{len(ranked)}\")\n", + " ranked_mrpc = sorted(results.items(),\n", + " key=lambda x: x[1][\"mrpc\"][\"f1\"], reverse=True)\n", + " pos = next(i for i, (n, _) in enumerate(ranked_mrpc) if n == \"captionbert\") + 1\n", + " print(f\" MRPC: #{pos}/{len(ranked_mrpc)}\")\n", + "\n", + " # vs best teacher\n", + " best_name = max((n for n in results if n != \"captionbert\"),\n", + " key=lambda n: results[n][\"stsb\"][\"spearman\"])\n", + " best_stsb = results[best_name][\"stsb\"][\"spearman\"]\n", + " student_stsb = results[\"captionbert\"][\"stsb\"][\"spearman\"]\n", + " print(f\"\\n vs Best teacher ({best_name}):\")\n", + " print(f\" STS-B gap: {student_stsb - best_stsb:+.4f}\")\n", + " print(f\" Param ratio: {results[best_name]['params'] / results['captionbert']['params']:.1f}×\")\n", + "\n", + " # Save\n", + " save_path = \"benchmark_captionbert_8192.json\"\n", + " with open(save_path, \"w\") as f:\n", + " json.dump(results, f, indent=2, default=str)\n", + " print(f\"\\n Saved to {save_path}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "8a0cd81be2be47888596acf176afddb9", + "84f657f86bd045359a1a1759d7194f76", + "dec09fc897a84c4189fdf2c4d26bc643", + "b02c30d0e2e44d9eb97699ea04616bc1", + "58509995a438455ba23ebe170493bc7e", + "3c06e0278e234c77ac77a750c2c5d578", + "fb7c3e3c8942451baf4ca758e9a6a946", + "d7e01fe40f6c4501875a99052345c3f0", + "5cb8b3e773b84a85ac577226e2857edc", + "4dd803a5e553472cbf00521945568d80", + "fff44aa0989441a596aa08b7bdf0b6b1", + "e99c0c2fd25c4688bde03694ea6fdc19", + "8871d95b022b415693cb2db9a3e7b775", + "a64c92181f864e41a9637e4ffd3cdb1c", + "30cdef2f92ba44eb9b42f442af3b6c42", + "6f9d36f1066b43d1986ce17cbfc95d9c", + "945d01f31faf4f929acfe4e99f9bde98", + "f4c52fb0719e4d38bbc1bc680117be78", + "2b1cf94985fc4440a22cc337f71f2f9c", + "122b374495c54486b721dea6f2ecf49a", + "0bacd6dcbe5f407dad12ff88ca0c734d", + "8e8d48014e514158ac40d1ca480dbe84", + "0e9d33c2824249bc9543659f9810e469", + "a02595570db14b8c949550d035f79d9e", + "17c946ace59444ec9b8504bcc96e7bb9", + "892fc7ea327d4bd29228cb577cf709ba", + "2646ff0b9bb941b3aeff743bc3c23ea9", + "0690fff8b38f4a44a07ad90232e7e727", + "003b40cfdf664f76958b2802917db0a3", + "a56949abd7b24bee8eb81586738815d7", + "748256ef83b94c0a9869503223dd2fcc", + "f2152636947d427bb30f8c8a77247dd8", + "6693c870f6494f40a3b64b021960d2b0", + "0921ba6b58d2433c82dac8fe3f2f4158", + "16a508809c504a8591adc161bd2c35b5", + "5c9e18ad9ecd480d876fcdf4cbf5d841", + "51a8fd04189b4cf488369e67c82588af", + "fb5ed0aaa3c34f47bf47c67472821422", + "bd2ea535cd6d432daf978af539e26344", + "2478bca9c91148268d1df4d27b0cdd72", + "6658e6702c974a6fa136992aa7fd0c90", + "9d3470ae0a0b4c93bcf93b604d70e64f", + "74e20f0f820b42f5875099ec103c6b6e", + "f0c276ad409f4cac9a52cc38825ba7da", + "3d89b978a63b4d4aa2b82cce48d99cc5", + "ee0532f1b4ee44c0a507878f40d0c2b9", + "cf7988090d4e4a32b48f80b4766993cf", + "f8a7d1fa879941eda3861daca4556620", + "21d612aac9f64fdaa29525fa13f34d07", + "e473894f50c04e70a4ae7b8cd34f1ac0", + "c93ad11d249b445b8941b5f628575fd6", + "bee5c523c01b40ea8cb919f13273cde1", + "e1e68445ab5c47c0ba6ee4636ad070a9", + "228b6eef214d4925960ee8bd32ef670e", + "6871542b5d374ee49e7082b35891f4dd" + ] + }, + "id": "Db4YL_S1j4tp", + "outputId": "71d43807-13b9-48ca-b95e-67222df66876" + }, + "execution_count": 2, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "BENCHMARK: geolip-captionbert-8192\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "LOADING BENCHMARKS\n", + "=================================================================\n", + " STS-B test: 1379 pairs\n", + " SICK-R test: 9927 pairs\n", + " MRPC test: 1725 pairs\n", + " Loading CC12M captions for retrieval (n=5000)...\n", + " Corpus: 4000, Queries: 1000\n", + "\n", + "=================================================================\n", + "LOADING: geolip-captionbert-8192\n", + "=================================================================\n", + " Downloaded: /root/.cache/huggingface/hub/models--AbstractPhil--geolip-captionbert-8192/snapshots/81a095cdf9d3f23cb03700ad5d7f14cfcaa74c35/best_model.pt\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/tmp/ipykernel_1319/2029101884.py:54: UserWarning: enable_nested_tensor is True, but self.use_nested_tensor is False because encoder_layer.norm_first was True\n", + " self.encoder = nn.TransformerEncoder(encoder_layer, num_layers=n_layers)\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " Parameters: 25,958,016\n", + " Encoding STS-B...\n", + " Encoding SICK-R...\n", + " Encoding MRPC...\n", + " Encoding captions...\n", + " STS-B: spearman=0.5032 pearson=0.5100\n", + " SICK-R: spearman=0.6138 pearson=0.6645\n", + " MRPC: f1=0.8068 acc=0.6881 thresh=0.71\n", + " Caption self-cos: mean=0.0040 max=0.7181\n", + "\n", + "=================================================================\n", + "EVALUATING: bert-base\n", + "=================================================================\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 1e-12]\n", + " entropy = -(S_pos * S_pos.log()).sum()\n", + "\n", + " trajectory.append({\n", + " \"spectrum\": S[:20].tolist(), # top 20 singular values\n", + " \"eff_dim\": eff_dim.item(),\n", + " \"entropy\": entropy.item(),\n", + " \"top1_ratio\": (S[0] / (S.sum() + 1e-12)).item(),\n", + " })\n", + "\n", + " results.append({\n", + " \"text\": data[\"texts\"][b],\n", + " \"trajectory\": trajectory,\n", + " })\n", + "\n", + " return results\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # 2. EFFECTIVE DIMENSIONALITY (output space)\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " def effective_dimensionality(self, data, k_neighbors=50):\n", + " \"\"\"\n", + " Local effective dimensionality around each embedding.\n", + " High = rich understanding. Low = surface-level placement.\n", + " \"\"\"\n", + " embeddings = data[\"final_embedding\"].float() # (B, 768)\n", + " B = embeddings.shape[0]\n", + "\n", + " if B < k_neighbors + 1:\n", + " k_neighbors = max(B - 1, 2)\n", + "\n", + " # Pairwise distances\n", + " sim = embeddings @ embeddings.T\n", + " results = []\n", + "\n", + " for b in range(B):\n", + " # Get k nearest neighbors\n", + " sims = sim[b].clone()\n", + " sims[b] = -1 # exclude self\n", + " _, topk_idx = sims.topk(k_neighbors)\n", + " neighbors = embeddings[topk_idx] # (k, 768)\n", + "\n", + " # Local PCA\n", + " centered = neighbors - neighbors.mean(0, keepdim=True)\n", + " try:\n", + " S = torch.linalg.svdvals(centered)\n", + " except Exception:\n", + " results.append({\"eff_dim\": 0, \"local_variance\": 0})\n", + " continue\n", + "\n", + " # Participation ratio\n", + " eff_dim = (S.sum() ** 2) / (S.pow(2).sum() + 1e-12)\n", + "\n", + " # How fast do eigenvalues decay?\n", + " S_norm = S / (S.sum() + 1e-12)\n", + " decay_rate = (S_norm[:5].sum() / S_norm.sum()).item()\n", + "\n", + " results.append({\n", + " \"text\": data[\"texts\"][b],\n", + " \"eff_dim\": eff_dim.item(),\n", + " \"decay_rate\": decay_rate, # high = concentrated, low = spread\n", + " \"local_spread\": centered.norm(dim=-1).mean().item(),\n", + " })\n", + "\n", + " return results\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # 3. CROSS-LAYER DIVERGENCE\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " def cross_layer_divergence(self, data):\n", + " \"\"\"\n", + " How much does the representation change between layers?\n", + " High change = computation happening. Low change = pass-through.\n", + " \"\"\"\n", + " results = []\n", + " n_layers = len(data[\"layer_pooled\"])\n", + " B = data[\"layer_pooled\"][0].shape[0]\n", + "\n", + " for b in range(B):\n", + " profile = []\n", + " for i in range(n_layers - 1):\n", + " h_curr = data[\"layer_pooled\"][i][b].float()\n", + " h_next = data[\"layer_pooled\"][i + 1][b].float()\n", + "\n", + " # Cosine between consecutive layers\n", + " cos = F.cosine_similarity(h_curr.unsqueeze(0),\n", + " h_next.unsqueeze(0)).item()\n", + " # L2 distance\n", + " l2 = (h_next - h_curr).norm().item()\n", + "\n", + " # Direction change (how much the direction rotates)\n", + " h_curr_n = F.normalize(h_curr, dim=0)\n", + " h_next_n = F.normalize(h_next, dim=0)\n", + " angle = torch.acos(torch.clamp(\n", + " (h_curr_n * h_next_n).sum(), -1, 1)).item()\n", + "\n", + " profile.append({\n", + " \"layer\": f\"{i}→{i+1}\",\n", + " \"cosine\": cos,\n", + " \"l2_shift\": l2,\n", + " \"angle_rad\": angle,\n", + " })\n", + "\n", + " # Total path length through representation space\n", + " total_path = sum(p[\"l2_shift\"] for p in profile)\n", + " # Where did most change happen?\n", + " max_shift_layer = max(range(len(profile)),\n", + " key=lambda i: profile[i][\"l2_shift\"])\n", + "\n", + " results.append({\n", + " \"text\": data[\"texts\"][b],\n", + " \"profile\": profile,\n", + " \"total_path\": total_path,\n", + " \"max_shift_layer\": max_shift_layer,\n", + " \"input_output_cos\": F.cosine_similarity(\n", + " data[\"layer_pooled\"][0][b].unsqueeze(0).float(),\n", + " data[\"layer_pooled\"][-1][b].unsqueeze(0).float()\n", + " ).item(),\n", + " })\n", + "\n", + " return results\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # 4. TOKEN INFLUENCE (gradient-based)\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " def token_influence(self, texts):\n", + " \"\"\"\n", + " Which tokens influence the output most?\n", + " Uses gradient of output norm w.r.t. input embeddings.\n", + " \"\"\"\n", + " if isinstance(texts, str):\n", + " texts = [texts]\n", + "\n", + " results = []\n", + " for text in texts:\n", + " inputs = self.tokenizer(\n", + " [text], max_length=self.max_len, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + "\n", + " # Get embedding layer output with gradients\n", + " input_ids = inputs[\"input_ids\"]\n", + " attention_mask = inputs[\"attention_mask\"]\n", + " n_real = attention_mask.sum().item()\n", + "\n", + " # Hook into embedding\n", + " emb = self.model.token_emb(input_ids) + \\\n", + " self.model.pos_emb(torch.arange(input_ids.shape[1],\n", + " device=DEVICE).unsqueeze(0))\n", + " emb = self.model.emb_drop(self.model.emb_norm(emb))\n", + " emb.retain_grad()\n", + "\n", + " # Forward through encoder\n", + " kpm = ~attention_mask.bool()\n", + " x = emb\n", + " for layer in self.model.encoder.layers:\n", + " x = layer(x, src_key_padding_mask=kpm)\n", + "\n", + " # Pool and project\n", + " mask = attention_mask.unsqueeze(-1).float()\n", + " pooled = (x * mask).sum(1) / mask.sum(1).clamp(min=1)\n", + " output = F.normalize(self.model.output_proj(pooled), dim=-1)\n", + "\n", + " # Gradient of output norm w.r.t embeddings\n", + " output.sum().backward()\n", + " grad = emb.grad[0].cpu()\n", + "\n", + " # Per-token influence = gradient norm\n", + " influence = grad.norm(dim=-1)[:int(n_real)] # only real tokens\n", + " influence = influence / (influence.sum() + 1e-12) # normalize\n", + "\n", + " # Decode tokens\n", + " token_ids = input_ids[0][:int(n_real)].cpu().tolist()\n", + " tokens = self.tokenizer.convert_ids_to_tokens(token_ids)\n", + "\n", + " results.append({\n", + " \"text\": text,\n", + " \"tokens\": tokens,\n", + " \"influence\": influence.tolist(),\n", + " \"top_tokens\": sorted(zip(tokens, influence.tolist()),\n", + " key=lambda x: -x[1])[:10],\n", + " \"concentration\": (influence.max() / influence.mean()).item(),\n", + " })\n", + "\n", + " self.model.zero_grad()\n", + "\n", + " return results\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # 5. FULL ANALYSIS\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " def analyze(self, texts):\n", + " \"\"\"Run all analyses on a set of texts.\"\"\"\n", + " if isinstance(texts, str):\n", + " texts = [texts]\n", + "\n", + " print(f\" Analyzing {len(texts)} inputs...\")\n", + "\n", + " data = self.extract_layers(texts)\n", + " spectral = self.spectral_trajectory(data)\n", + " eff_dim = self.effective_dimensionality(data)\n", + " divergence = self.cross_layer_divergence(data)\n", + " influence = self.token_influence(texts)\n", + "\n", + " report = {}\n", + " for i, text in enumerate(texts):\n", + " report[text] = {\n", + " \"embedding\": data[\"final_embedding\"][i],\n", + " \"n_tokens\": data[\"n_tokens\"][i].item(),\n", + " \"spectral\": spectral[i],\n", + " \"eff_dim\": eff_dim[i] if i < len(eff_dim) else {},\n", + " \"divergence\": divergence[i],\n", + " \"influence\": influence[i],\n", + " }\n", + "\n", + " return report\n", + "\n", + " # ══════��═══════════════════════════════════════════════════════\n", + " # PRINTING\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " def print_report(self, report):\n", + " \"\"\"Print full analysis report.\"\"\"\n", + " print(f\"\\n{'='*70}\")\n", + " print(\"INTERNAL ANALYSIS REPORT\")\n", + " print(f\"{'='*70}\")\n", + "\n", + " # Summary table\n", + " print(f\"\\n {'Text':<25} {'Tokens':>6} {'EffDim':>7} {'Path':>7} \"\n", + " f\"{'MaxShift':>9} {'InOutCos':>8} {'Concentrate':>11}\")\n", + " print(f\" {'-'*75}\")\n", + "\n", + " for text, r in report.items():\n", + " label = text[:24]\n", + " ed = r[\"eff_dim\"].get(\"eff_dim\", 0)\n", + " tp = r[\"divergence\"][\"total_path\"]\n", + " ms = r[\"divergence\"][\"max_shift_layer\"]\n", + " ioc = r[\"divergence\"][\"input_output_cos\"]\n", + " conc = r[\"influence\"][\"concentration\"]\n", + " print(f\" {label:<25} {r['n_tokens']:>6} {ed:>7.1f} {tp:>7.2f} \"\n", + " f\" layer {ms:>2} {ioc:>7.3f} {conc:>10.1f}\")\n", + "\n", + " # Spectral evolution\n", + " print(f\"\\n SPECTRAL TRAJECTORY (effective dim per layer):\")\n", + " print(f\" {'Text':<25}\", end=\"\")\n", + " n_layers = len(next(iter(report.values()))[\"spectral\"][\"trajectory\"])\n", + " for i in range(n_layers):\n", + " print(f\" L{i:>2}\", end=\"\")\n", + " print()\n", + " print(f\" {'-'*75}\")\n", + "\n", + " for text, r in report.items():\n", + " label = text[:24]\n", + " print(f\" {label:<25}\", end=\"\")\n", + " for step in r[\"spectral\"][\"trajectory\"]:\n", + " ed = step.get(\"eff_dim\", 0)\n", + " print(f\" {ed:>4.0f}\", end=\"\")\n", + " print()\n", + "\n", + " # Spectral entropy per layer\n", + " print(f\"\\n SPECTRAL ENTROPY (information content per layer):\")\n", + " print(f\" {'Text':<25}\", end=\"\")\n", + " for i in range(n_layers):\n", + " print(f\" L{i:>2}\", end=\"\")\n", + " print()\n", + " print(f\" {'-'*75}\")\n", + "\n", + " for text, r in report.items():\n", + " label = text[:24]\n", + " print(f\" {label:<25}\", end=\"\")\n", + " for step in r[\"spectral\"][\"trajectory\"]:\n", + " ent = step.get(\"entropy\", 0)\n", + " print(f\" {ent:>4.1f}\", end=\"\")\n", + " print()\n", + "\n", + " # Cross-layer divergence profiles\n", + " print(f\"\\n COMPUTATION PROFILE (L2 shift between layers):\")\n", + " print(f\" {'Text':<25}\", end=\"\")\n", + " for i in range(n_layers - 1):\n", + " print(f\" {i}→{i+1:>2}\", end=\"\")\n", + " print()\n", + " print(f\" {'-'*75}\")\n", + "\n", + " for text, r in report.items():\n", + " label = text[:24]\n", + " print(f\" {label:<25}\", end=\"\")\n", + " for step in r[\"divergence\"][\"profile\"]:\n", + " print(f\" {step['l2_shift']:>4.1f}\", end=\"\")\n", + " print()\n", + "\n", + " # Token influence for each input\n", + " print(f\"\\n TOKEN INFLUENCE (top contributing tokens):\")\n", + " for text, r in report.items():\n", + " top = r[\"influence\"][\"top_tokens\"][:5]\n", + " tok_str = \" \".join(f\"{t}={v:.3f}\" for t, v in top)\n", + " print(f\" {text[:40]:<42} {tok_str}\")\n", + "\n", + " def compare(self, report, text_a, text_b):\n", + " \"\"\"Compare internal representations of two specific inputs.\"\"\"\n", + " a = report[text_a]\n", + " b = report[text_b]\n", + "\n", + " cos = F.cosine_similarity(\n", + " a[\"embedding\"].unsqueeze(0),\n", + " b[\"embedding\"].unsqueeze(0)).item()\n", + "\n", + " print(f\"\\n{'='*70}\")\n", + " print(f\"COMPARISON: '{text_a}' vs '{text_b}'\")\n", + " print(f\"{'='*70}\")\n", + " print(f\" Output cosine: {cos:.4f}\")\n", + " print(f\" Tokens: {a['n_tokens']} vs {b['n_tokens']}\")\n", + "\n", + " # Effective dim comparison\n", + " ed_a = a[\"eff_dim\"].get(\"eff_dim\", 0)\n", + " ed_b = b[\"eff_dim\"].get(\"eff_dim\", 0)\n", + " print(f\" Effective dim: {ed_a:.1f} vs {ed_b:.1f} (Δ={abs(ed_a-ed_b):.1f})\")\n", + "\n", + " # Path comparison\n", + " pa = a[\"divergence\"][\"total_path\"]\n", + " pb = b[\"divergence\"][\"total_path\"]\n", + " print(f\" Total path: {pa:.2f} vs {pb:.2f} (Δ={abs(pa-pb):.2f})\")\n", + "\n", + " # Layer-by-layer spectral comparison\n", + " print(f\"\\n Effective dim trajectory:\")\n", + " print(f\" {'Layer':<8} {'A':>8} {'B':>8} {'Δ':>8}\")\n", + " traj_a = a[\"spectral\"][\"trajectory\"]\n", + " traj_b = b[\"spectral\"][\"trajectory\"]\n", + " for i in range(len(traj_a)):\n", + " ea = traj_a[i].get(\"eff_dim\", 0)\n", + " eb = traj_b[i].get(\"eff_dim\", 0)\n", + " print(f\" L{i:<6} {ea:>8.1f} {eb:>8.1f} {abs(ea-eb):>8.1f}\")\n", + "\n", + " # Divergence profile comparison\n", + " print(f\"\\n Computation profile (L2 shift):\")\n", + " print(f\" {'Transition':<10} {'A':>8} {'B':>8} {'Δ':>8}\")\n", + " for i in range(len(a[\"divergence\"][\"profile\"])):\n", + " sa = a[\"divergence\"][\"profile\"][i][\"l2_shift\"]\n", + " sb = b[\"divergence\"][\"profile\"][i][\"l2_shift\"]\n", + " label = a[\"divergence\"][\"profile\"][i][\"layer\"]\n", + " print(f\" {label:<10} {sa:>8.2f} {sb:>8.2f} {abs(sa-sb):>8.2f}\")\n", + "\n", + " # Token influence comparison\n", + " print(f\"\\n Top tokens:\")\n", + " print(f\" A: {' '.join(f'{t}={v:.3f}' for t,v in a['influence']['top_tokens'][:5])}\")\n", + " print(f\" B: {' '.join(f'{t}={v:.3f}' for t,v in b['influence']['top_tokens'][:5])}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# RUN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "if __name__ == \"__main__\":\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " REPO_ID = \"AbstractPhil/geolip-captionbert-8192\"\n", + " print(\"Loading model...\")\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True)\n", + " tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + "\n", + " analyzer = InternalAnalyzer(model, tokenizer)\n", + "\n", + " # Test words spanning known-domain and unknown-domain\n", + " test_words = [\n", + " # Known domain (captions)\n", + " #\"girl\",\n", + " #\"woman\",\n", + " #\"dog\",\n", + " #\"sunset\",\n", + " #\"painting\",\n", + " ## Unknown domain (abstract)\n", + " #\"subtraction\",\n", + " #\"multiplication\",\n", + " #\"prophetic\",\n", + " #\"differential\",\n", + " #\"adjacency\",\n", + " ## Phrases\n", + " #\"a girl sitting near a window\",\n", + " #\"a dog playing on the beach\",\n", + " #\"the differential equation of motion\",\n", + " \"a potato on top of a table\",\n", + " \"a table on top of a potato\",\n", + " ]\n", + "\n", + " report = analyzer.analyze(test_words)\n", + " analyzer.print_report(report)\n", + "\n", + " # Direct comparisons\n", + " analyzer.compare(report, \"a potato on top of a table\", \"a table on top of a potato\")\n", + " #analyzer.compare(report, \"girl\", \"subtraction\")\n", + " #analyzer.compare(report, \"a girl sitting near a window\",\n", + " #\"the differential equation of motion\")\n", + "\n", + " print(f\"\\n{'='*70}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*70}\")" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "fc4154afb1d54ade92bd8f1f03d85d22", + "242d8f5bd00b4a8a82d13cab2fedbb11", + "5d6c91de29bd4689b951e8f8da31723a", + "aadae893f97046449cfdb92fe54619f1", + "dd2b7d9da2044054b13463ef08789981", + "e3ac055fbe0a476ba92c9ac428c46d3c", + "56dc19f3b096416a84a308c5ab866578", + "e2e5e2362d904f108884756cda1d6518", + "0d655e4f97c24655a08001461a8824c2", + "1019f3ce8439413b975e1ff6444cf61a", + "e4b69e37104d4a60877e9d29062baea1" + ] + }, + "id": "xxPw0F4OAfDA", + "outputId": "4aa82337-554f-45a7-9ae7-d53aa3e79aa7" + }, + "execution_count": 10, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Loading model...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/82 [00:00= 0)\n", + " val_ds = ds[\"validation\"].filter(lambda x: x[\"label\"] >= 0)\n", + " print(f\" Train: {len(train_ds):,} Val: {len(val_ds):,}\")\n", + "\n", + " # Pre-encode all examples (frozen backbone, only need to do once)\n", + " print(\"\\nPre-encoding with frozen backbone...\")\n", + "\n", + " @torch.no_grad()\n", + " def encode_pairs(dataset, max_n=None, batch_size=256):\n", + " if max_n:\n", + " dataset = dataset.select(range(min(max_n, len(dataset))))\n", + "\n", + " all_p_tokens, all_h_tokens = [], []\n", + " all_p_pooled, all_h_pooled = [], []\n", + " all_p_mask, all_h_mask = [], []\n", + " all_labels = []\n", + "\n", + " for i in tqdm(range(0, len(dataset), batch_size), desc=\" Encoding\"):\n", + " j = min(i + batch_size, len(dataset))\n", + " batch = dataset[i:j]\n", + "\n", + " # Premise\n", + " p_inputs = tokenizer(\n", + " batch[\"premise\"], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_inputs)\n", + "\n", + " # Hypothesis\n", + " h_inputs = tokenizer(\n", + " batch[\"hypothesis\"], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " h_out = model(**h_inputs)\n", + "\n", + " all_p_tokens.append(p_out.token_embeddings.cpu())\n", + " all_h_tokens.append(h_out.token_embeddings.cpu())\n", + " all_p_pooled.append(p_out.last_hidden_state.cpu())\n", + " all_h_pooled.append(h_out.last_hidden_state.cpu())\n", + " all_p_mask.append(p_inputs[\"attention_mask\"].cpu())\n", + " all_h_mask.append(h_inputs[\"attention_mask\"].cpu())\n", + " all_labels.append(torch.tensor(batch[\"label\"]))\n", + "\n", + " return {\n", + " \"p_tokens\": torch.cat(all_p_tokens),\n", + " \"h_tokens\": torch.cat(all_h_tokens),\n", + " \"p_pooled\": torch.cat(all_p_pooled),\n", + " \"h_pooled\": torch.cat(all_h_pooled),\n", + " \"p_mask\": torch.cat(all_p_mask),\n", + " \"h_mask\": torch.cat(all_h_mask),\n", + " \"labels\": torch.cat(all_labels),\n", + " }\n", + "\n", + " train_data = encode_pairs(train_ds, max_n=100000)\n", + " val_data = encode_pairs(val_ds, max_n=10000)\n", + "\n", + " # Free backbone from GPU\n", + " del model\n", + " torch.cuda.empty_cache()\n", + " import gc; gc.collect()\n", + "\n", + " # Build NLI head\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"NLI HEAD\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli = NLIHead(d_token=384, d_pooled=768, n_heads=6,\n", + " n_cross_layers=2, n_classes=3, dropout=0.1).to(DEVICE)\n", + " n_params = sum(p.numel() for p in nli.parameters())\n", + " print(f\" Parameters: {n_params:,}\")\n", + "\n", + " # Training setup\n", + " EPOCHS = 15\n", + " BATCH_SIZE = 128\n", + " LR = 1e-4 # lower LR — drift loss constrains optimization\n", + "\n", + " optimizer = torch.optim.AdamW(nli.parameters(), lr=LR, weight_decay=0.01)\n", + " n_train = train_data[\"labels\"].shape[0]\n", + " n_batches = n_train // BATCH_SIZE\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * EPOCHS, eta_min=1e-6)\n", + "\n", + " print(f\" Epochs: {EPOCHS}\")\n", + " print(f\" Batch size: {BATCH_SIZE}\")\n", + " print(f\" Batches/epoch: {n_batches}\")\n", + "\n", + " # Move pre-encoded data to GPU\n", + " for k in train_data:\n", + " train_data[k] = train_data[k].to(DEVICE)\n", + " for k in val_data:\n", + " val_data[k] = val_data[k].to(DEVICE)\n", + "\n", + " # Train\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({EPOCHS} epochs)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " best_val_acc = 0.0\n", + " for epoch in range(EPOCHS):\n", + " nli.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss, total_correct, n = 0, 0, 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BATCH_SIZE):\n", + " idx = perm[i:i+BATCH_SIZE]\n", + " if len(idx) < 4:\n", + " continue\n", + "\n", + " logits, aux = nli(\n", + " train_data[\"p_tokens\"][idx],\n", + " train_data[\"h_tokens\"][idx],\n", + " train_data[\"p_pooled\"][idx],\n", + " train_data[\"h_pooled\"][idx],\n", + " train_data[\"p_mask\"][idx],\n", + " train_data[\"h_mask\"][idx],\n", + " )\n", + " labels = train_data[\"labels\"][idx]\n", + "\n", + " # ── Full geometric loss ──\n", + " # 1. Task loss\n", + " l_ce = F.cross_entropy(logits, labels)\n", + "\n", + " # 2. Manifold drift — stay on backbone manifold\n", + " l_drift = aux[\"manifold_drift\"]\n", + "\n", + " # 3. InfoNCE — cross-attended P should retrieve its own H\n", + " l_nce = infonce(aux[\"p_cross_pooled\"], aux[\"h_cross_pooled\"])\n", + "\n", + " # 4. Pentachoron CV — cross-attended space stays geometrically regular\n", + " cross_embs = torch.cat([aux[\"p_cross_pooled\"], aux[\"h_cross_pooled\"]], dim=0)\n", + " l_cv = cv_loss(F.normalize(cross_embs, dim=-1), target=0.084)\n", + "\n", + " # 5. Gate floor — prevent gates from collapsing to zero\n", + " gate_floor = 0.05\n", + " l_gate = (F.relu(gate_floor - nli.p_gate) +\n", + " F.relu(gate_floor - nli.h_gate))\n", + "\n", + " loss = (0.5 * l_ce +\n", + " 1.0 * l_drift +\n", + " 1.0 * l_nce +\n", + " 0.1 * l_cv +\n", + " 1.0 * l_gate)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(nli.parameters(), 1.0)\n", + " optimizer.step()\n", + " optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " total_loss += l_ce.item()\n", + " total_correct += (logits.argmax(-1) == labels).sum().item()\n", + " n += len(idx)\n", + "\n", + " elapsed = time.time() - t0\n", + " train_acc = total_correct / max(n, 1)\n", + " train_loss = total_loss / max(n // BATCH_SIZE, 1)\n", + "\n", + " # ── Full Geometric Validation ──\n", + " nli.eval()\n", + " with torch.no_grad():\n", + " val_n = val_data[\"labels\"].shape[0]\n", + " val_correct = 0\n", + " val_loss = 0\n", + " val_drift = 0\n", + " all_p_cross = []\n", + " all_h_cross = []\n", + " all_logits = []\n", + " all_labels = []\n", + "\n", + " for i in range(0, val_n, 512):\n", + " j = min(i + 512, val_n)\n", + " logits, aux = nli(\n", + " val_data[\"p_tokens\"][i:j],\n", + " val_data[\"h_tokens\"][i:j],\n", + " val_data[\"p_pooled\"][i:j],\n", + " val_data[\"h_pooled\"][i:j],\n", + " val_data[\"p_mask\"][i:j],\n", + " val_data[\"h_mask\"][i:j],\n", + " )\n", + " labels = val_data[\"labels\"][i:j]\n", + " val_correct += (logits.argmax(-1) == labels).sum().item()\n", + " val_loss += F.cross_entropy(logits, labels, reduction=\"sum\").item()\n", + " val_drift += aux[\"manifold_drift\"].item()\n", + " all_p_cross.append(aux[\"p_cross_pooled\"].cpu())\n", + " all_h_cross.append(aux[\"h_cross_pooled\"].cpu())\n", + " all_logits.append(logits.cpu())\n", + " all_labels.append(labels.cpu())\n", + "\n", + " val_acc = val_correct / val_n\n", + " val_loss = val_loss / val_n\n", + " val_drift = val_drift / max(val_n // 512, 1)\n", + "\n", + " # ── Geometric measurements ──\n", + " p_cross_all = torch.cat(all_p_cross).to(DEVICE)\n", + " h_cross_all = torch.cat(all_h_cross).to(DEVICE)\n", + " logits_all = torch.cat(all_logits)\n", + " labels_all = torch.cat(all_labels)\n", + "\n", + " # Pentachoron CV on cross-attended space\n", + " cross_embs = F.normalize(\n", + " torch.cat([p_cross_all[:1000], h_cross_all[:1000]], dim=0), dim=-1)\n", + " B_cv = cross_embs.shape[0]\n", + " vols = []\n", + " for _ in range(200):\n", + " idx = torch.randperm(B_cv, device=DEVICE)[:5]\n", + " v2 = cayley_menger_vol2(cross_embs[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0:\n", + " vols.append(v)\n", + " vols_arr = np.array(vols)\n", + " cross_cv = float(vols_arr.std() / (vols_arr.mean() + 1e-8)) if len(vols) > 10 else 0.0\n", + "\n", + " # Cross-attended P↔H cosine (should be high for entailment, low for contradiction)\n", + " ph_cos = F.cosine_similarity(p_cross_all[:2000], h_cross_all[:2000], dim=-1)\n", + " ent_mask = labels_all[:2000] == 0\n", + " con_mask = labels_all[:2000] == 2\n", + " neu_mask = labels_all[:2000] == 1\n", + " cos_ent = ph_cos[ent_mask].mean().item() if ent_mask.sum() > 0 else 0\n", + " cos_con = ph_cos[con_mask].mean().item() if con_mask.sum() > 0 else 0\n", + " cos_neu = ph_cos[neu_mask].mean().item() if neu_mask.sum() > 0 else 0\n", + "\n", + " # Effective dimensionality of cross-attended space\n", + " centered = cross_embs[:2000] - cross_embs[:2000].mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered.float())\n", + " eff_dim = (S.sum() ** 2) / (S.pow(2).sum() + 1e-12)\n", + "\n", + " # Per-class accuracy (full val set)\n", + " preds = logits_all.argmax(-1)\n", + " ent_mask_full = labels_all == 0\n", + " con_mask_full = labels_all == 2\n", + " neu_mask_full = labels_all == 1\n", + " acc_ent = (preds[ent_mask_full] == 0).float().mean().item() if ent_mask_full.sum() > 0 else 0\n", + " acc_con = (preds[con_mask_full] == 2).float().mean().item() if con_mask_full.sum() > 0 else 0\n", + " acc_neu = (preds[neu_mask_full] == 1).float().mean().item() if neu_mask_full.sum() > 0 else 0\n", + "\n", + " del p_cross_all, h_cross_all, cross_embs\n", + "\n", + " pg = nli.p_gate.item()\n", + " hg = nli.h_gate.item()\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s\")\n", + " print(f\" Task: loss={train_loss:.4f} t_acc={train_acc:.4f} v_acc={val_acc:.4f} v_loss={val_loss:.4f}\")\n", + " print(f\" Geometry: CV={cross_cv:.4f} eff_dim={eff_dim:.1f} drift={val_drift:.5f}\")\n", + " print(f\" Gates: p={pg:.4f} h={hg:.4f}\")\n", + " print(f\" P↔H cos: ent={cos_ent:.3f} neu={cos_neu:.3f} con={cos_con:.3f} (spread={cos_ent-cos_con:.3f})\")\n", + " print(f\" Per-class: ent={acc_ent:.3f} neu={acc_neu:.3f} con={acc_con:.3f}\")\n", + "\n", + " if val_acc > best_val_acc:\n", + " best_val_acc = val_acc\n", + " torch.save(nli.state_dict(), \"nli_head_best.pt\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # COMPOSITIONAL TEST\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL ORDER TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli.load_state_dict(torch.load(\"nli_head_best.pt\", weights_only=True))\n", + " nli.eval()\n", + "\n", + " # Reload backbone for test\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True)\n", + " model = model.to(DEVICE).eval()\n", + "\n", + " label_names = [\"entailment\", \"neutral\", \"contradiction\"]\n", + "\n", + " test_pairs = [\n", + " (\"a potato on top of a table\", \"a table on top of a potato\"),\n", + " (\"a potato on top of a table\", \"there is a potato\"),\n", + " (\"a cat is sitting on a mat\", \"a mat is sitting on a cat\"),\n", + " (\"a dog chased the cat\", \"the cat chased the dog\"),\n", + " (\"a woman is holding a baby\", \"a baby is holding a woman\"),\n", + " (\"the boy kicked the ball\", \"the ball kicked the boy\"),\n", + " (\"a man is riding a horse\", \"a horse is riding a man\"),\n", + " ]\n", + "\n", + " with torch.no_grad():\n", + " for premise, hypothesis in test_pairs:\n", + " p_in = tokenizer([premise], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " h_in = tokenizer([hypothesis], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + " h_out = model(**h_in)\n", + "\n", + " logits, _ = nli(\n", + " p_out.token_embeddings, h_out.token_embeddings,\n", + " p_out.last_hidden_state, h_out.last_hidden_state,\n", + " p_in[\"attention_mask\"], h_in[\"attention_mask\"],\n", + " )\n", + " probs = F.softmax(logits, dim=-1)[0]\n", + " pred = label_names[probs.argmax()]\n", + "\n", + " # Also show pooled cosine for comparison\n", + " cos = F.cosine_similarity(\n", + " p_out.last_hidden_state, h_out.last_hidden_state).item()\n", + "\n", + " print(f\"\\n P: {premise}\")\n", + " print(f\" H: {hypothesis}\")\n", + " print(f\" Pooled cos: {cos:.3f} (order-blind)\")\n", + " print(f\" NLI: {pred} \"\n", + " f\"[E={probs[0]:.3f} N={probs[1]:.3f} C={probs[2]:.3f}]\")\n", + "\n", + " print(f\"\\n Best val accuracy: {best_val_acc:.4f}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " train_nli_head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "e1a8d24aa52a43ce816a516d858fdf66", + "55ced514f9894c65bcfb78f8f64b72b9", + "961c1273295040a98df4bcb424d83ef0", + "082caf846d1e4ffa8f10e99d4d1ef285", + "e7365eb9ad9449e8a34d9ebcaa5590f4", + "3ca334c988de406fab7c123a28188fba", + "f1585bd61f234038a0482ce2a64815c1", + "2aea60dac24e4bb7a11a3d95e1420a93", + "b8190678a51f4e41ab1a9b9625c3a9e6", + "bd0484aa4a96450d8d820bddf38bb36e", + "f517d583ed7046dfb6fa9c03d355d034", + "9ce5f076aa2c4cab9375fb3d7a4289b1", + "7b6c598e02e7484b90b0c7178a82a2fc", + "4ad5d6904a0348fea992da43585b00ef", + "249ff1a4e6b94a9185642e20c5f1bda8", + "b13bbe61c7804367b740463d6f24ac3b", + "413f5527831c49829a96e0ac303d4a1a", + "fb1f6e367eec4cdbb1be205bb76b7b83", + "845d21a05b404a1aae834514a3bdddfa", + "ecc945a6b0e344548301f6f2b3477844", + "37e0ad2f47074b4795a717a64a22a254", + "2018eec3287f46a7a2ae822f560e6f96" + ] + }, + "id": "myB0A8FEWtdE", + "outputId": "2ce604f5-00ea-4b40-f712-9b4579612a41" + }, + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "NLI HEAD TRAINING\n", + "=================================================================\n", + "\n", + "Loading backbone...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/82 [00:00= 10:\n", + " ctx_n = F.normalize(geo_context, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = ctx_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " bank_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"bank_cv\"] = bank_cv\n", + " else:\n", + " aux[\"bank_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # Summary diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"anchor_mean_cos\"] = anchor_cos.mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux, cv_target=0.15):\n", + " \"\"\"Combined bank training loss.\"\"\"\n", + " loss = (1.0 * aux[\"expert_agreement\"] +\n", + " 1.0 * aux[\"rotation_ortho\"] +\n", + " 0.5 * aux[\"anchor_spread\"] +\n", + " 0.1 * aux[\"anchor_entropy\"] +\n", + " 0.3 * (aux[\"bank_cv\"] - cv_target).abs())\n", + " return loss\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.12, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION + ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float()\n", + " T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True)\n", + " t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean\n", + " N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]\n", + " x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]\n", + " return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " N_SAMPLES = 20000\n", + " MAX_LEN = 128\n", + " BATCH = 256\n", + "\n", + " # ── Phase 0: Extract ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXTRACTION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_SAMPLES:\n", + " break\n", + " print(f\" Captions: {len(captions):,}\")\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " print(f\"\\n Extracting: {short}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + " embeds[short] = torch.cat(all_emb)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 0b: Align + Consensus ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0b: PROCRUSTES ALIGNMENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref = \"bert\"\n", + " names = [s for _, s, _ in EXPERTS]\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], embeds[ref])\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " print(f\" {name:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}\")\n", + "\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " print(f\" Consensus: {consensus.shape}\")\n", + " for name in names:\n", + " cos = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {cos:.4f}\")\n", + "\n", + " consensus_cv = cv_metric(consensus[:2000].to(DEVICE))\n", + " print(f\" Consensus CV: {consensus_cv:.4f}\")\n", + "\n", + " del embeds, aligned\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 1: Train Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: TRAIN STUDENT (2 experts, 20K captions)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " tokens = tokenizer(captions, max_length=MAX_LEN, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N_SAMPLES - 2000\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " student = MiniStudent(\n", + " vocab_size=tokenizer.vocab_size, max_len=MAX_LEN,\n", + " d_model=256, n_heads=4, n_layers=4, d_ff=1024,\n", + " output_dim=768, dropout=0.1, pad_token_id=tokenizer.pad_token_id\n", + " ).to(DEVICE)\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Student: {n_params:,} params\")\n", + "\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=3e-4, weight_decay=0.01)\n", + "\n", + " for epoch in range(5):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " t_loss, t_acc, t_cos, n = 0, 0, 0, 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_cv)\n", + " loss = l_nce + l_mse + 0.1 * l_cv\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " t_loss += loss.item(); t_acc += acc; t_cos += cos; n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + " student.eval()\n", + " with torch.no_grad():\n", + " v_emb = student(val_ids, val_mask)\n", + " _, v_acc = infonce(v_emb[:1000], val_targets[:1000])\n", + " v_cos = F.cosine_similarity(v_emb, val_targets, dim=-1).mean().item()\n", + " v_cv = cv_metric(v_emb[:1000])\n", + "\n", + " print(f\" E{epoch+1}: {elapsed:.0f}s loss={t_loss/d:.4f} \"\n", + " f\"t_acc={t_acc/d:.3f} t_cos={t_cos/d:.3f} \"\n", + " f\"v_acc={v_acc:.3f} v_cos={v_cos:.3f} v_cv={v_cv:.3f}\")\n", + "\n", + " # Save student\n", + " torch.save(student.state_dict(), \"mini_student.pt\")\n", + " print(f\"\\n Student saved. v_cos={v_cos:.3f}, v_cv={v_cv:.3f}\")\n", + "\n", + " # ── Phase 2: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: TRAIN ALIGNMENT BANK (student frozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " # Freeze student\n", + " student.eval()\n", + " for p in student.parameters():\n", + " p.requires_grad = False\n", + "\n", + " # Pre-encode everything through frozen student\n", + " print(\" Pre-encoding through frozen student...\")\n", + " with torch.no_grad():\n", + " all_embs = []\n", + " for i in range(0, n_train, 512):\n", + " j = min(i + 512, n_train)\n", + " emb = student(train_ids[i:j], train_mask[i:j])\n", + " all_embs.append(emb)\n", + " student_embs = torch.cat(all_embs) # (n_train, 768)\n", + " val_student_embs = student(val_ids, val_mask)\n", + "\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " # Build bank\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=512, d_bank=64\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names, consensus[:n_train])\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=1e-3, weight_decay=0.01)\n", + " BANK_EPOCHS = 20\n", + " BANK_BATCH = 256\n", + "\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0,\n", + " \"anchor_spread\": 0, \"bank_cv\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + "\n", + " emb = student_embs[idx]\n", + " enriched, aux = bank(emb)\n", + " loss = bank.bank_loss(aux, cv_target=consensus_cv + 0.02)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + "\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0\n", + " d = max(n, 1)\n", + "\n", + " # Validation\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " v_enriched, v_aux = bank(val_student_embs)\n", + " v_loss = bank.bank_loss(v_aux, cv_target=consensus_cv + 0.02).item()\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} \"\n", + " f\"v_loss={v_loss:.4f} \"\n", + " f\"expert_agr={stats['expert_agreement']/d:.5f} \"\n", + " f\"ortho={stats['rotation_ortho']/d:.5f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} \"\n", + " f\"cv={stats['bank_cv']/d:.4f} \"\n", + " f\"anchor_max={v_aux['anchor_max_cos']:.3f} \"\n", + " f\"expert_cos={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank.pt\")\n", + "\n", + " # ── Phase 3: Verify Geometry ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " # Check that enriched embeddings preserve original structure\n", + " enriched_val, _ = bank(val_student_embs)\n", + " original_768 = enriched_val[:, :768] # first 768 dims = original embedding\n", + " geo_context = enriched_val[:, 768:] # last d_bank dims = geometric annotation\n", + "\n", + " # Original embedding should be unchanged (passthrough)\n", + " passthrough_cos = F.cosine_similarity(\n", + " original_768[:100], val_student_embs[:100], dim=-1).mean().item()\n", + "\n", + " # Geometric context should be informative\n", + " geo_cv = cv_metric(F.normalize(geo_context[:1000], dim=-1))\n", + " geo_eff_dim = torch.linalg.svdvals(\n", + " geo_context[:1000].float() - geo_context[:1000].float().mean(0)).pow(2)\n", + " geo_eff_dim = (geo_eff_dim.sum() ** 2) / (geo_eff_dim.pow(2).sum() + 1e-12)\n", + "\n", + " print(f\" Passthrough integrity: {passthrough_cos:.6f} (should be ~1.000)\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo context eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Geo context shape: {geo_context.shape}\")\n", + "\n", + " # ── Phase 4: Quick Classifier Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " # Create synthetic 3-class task from similarity structure\n", + " # Class 0: high consensus cosine pairs (similar)\n", + " # Class 1: medium consensus cosine pairs\n", + " # Class 2: low consensus cosine pairs (different)\n", + " with torch.no_grad():\n", + " # Generate synthetic labels from embedding distances\n", + " embs = val_student_embs[:1000]\n", + " sim = embs @ embs.T\n", + " sim.fill_diagonal_(-1) # exclude self\n", + "\n", + " # Random pairs\n", + " n_pairs = 3000\n", + " idx_a = torch.randint(0, 1000, (n_pairs,))\n", + " idx_b = torch.randint(0, 1000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + "\n", + " # Assign labels by cosine terciles\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0 # similar\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1 # medium\n", + " labels[pair_cos <= t1] = 2 # different\n", + "\n", + " # Get enriched representations\n", + " enriched_a, _ = bank(embs[idx_a])\n", + " enriched_b, _ = bank(embs[idx_b])\n", + "\n", + " # Train tiny classifier: with bank vs without bank\n", + " for mode in [\"with_bank\", \"without_bank\"]:\n", + " if mode == \"with_bank\":\n", + " feat_dim = (768 + 64) * 2 # enriched\n", + " features = torch.cat([enriched_a, enriched_b], dim=-1)\n", + " else:\n", + " feat_dim = 768 * 2 # raw\n", + " features = torch.cat([embs[idx_a], embs[idx_b]], dim=-1)\n", + "\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, 128), nn.GELU(),\n", + " nn.Linear(128, 3)\n", + " ).to(DEVICE)\n", + "\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " n_clf_train = 2400\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + "\n", + " for e in range(20):\n", + " clf.train()\n", + " logits = clf(train_f)\n", + " loss = F.cross_entropy(logits, train_l)\n", + " loss.backward()\n", + " clf_opt.step(); clf_opt.zero_grad()\n", + "\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " val_logits = clf(val_f)\n", + " val_acc = (val_logits.argmax(-1) == val_l).float().mean().item()\n", + " train_logits = clf(train_f)\n", + " train_acc = (train_logits.argmax(-1) == train_l).float().mean().item()\n", + "\n", + " print(f\" {mode:15s}: train_acc={train_acc:.3f} val_acc={val_acc:.3f} \"\n", + " f\"gap={train_acc-val_acc:.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + " print(f\"\\n Student: mini_student.pt\")\n", + " print(f\" Bank: alignment_bank.pt\")\n", + " print(f\" Consensus CV: {consensus_cv:.4f}\")\n", + " print(f\" Student v_cos: {v_cos:.3f}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "b7f0462796534c0c887869dd4ee536b6", + "b04346b7cdd44d1ca83bb3f9164d34b7", + "4c0fd3ff39ec45029ba60e6cef15cd2b", + "08611fc88d5143e6b79234aaae1886ef", + "5420ef57acbd4189832bce4f46fb3c78", + "e6d9053451594834970c6a48833acbe0", + "0917a8224cb645aeaa9beefd8dddcdda", + "9027d19005a5449189ec7a1440f3d9b7", + "798c1ea91e06483389133dc069842676", + "66168ca2364347aab0174721ab52e400", + "621c518987f74847a99d25597315e17c", + "0074e789f07c4864a2d0bdae43a0a950", + "fa510b450a2e4b8d929cf23ba0ce8e6f", + "352f5aa1999143e1840fd185b76a1980", + "4884f934d3be4876b99112e5d9cc42ba", + "e45c8496bebd491ca4343026f860f6f6", + "1560366419274f78882efd9df291137b", + "262abd0b7042430ab85f64c184f21fc5", + "7235412109f54367a9d19b29b8481df6", + "062b0957ba7e49bb9fb449da603061ee", + "0638d340162c4f0aac6d78f0d7856d59", + "28cc211bf2194de3a931c0f637d8edc9" + ] + }, + "id": "2P5nrXApt7Ls", + "outputId": "988d30e6-2912-463f-b0d1-edd0d7e0103a" + }, + "execution_count": 1, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "RAPID PROTOTYPE: 2-Expert Consensus + Alignment Bank\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "PHASE 0: EXTRACTION\n", + "=================================================================\n", + " Captions: 20,000\n", + "\n", + " Extracting: bert...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 0:\n", + " aux[\"cross_expert_var\"] = cross_features.var(dim=0).mean()\n", + " else:\n", + " aux[\"cross_expert_var\"] = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # 6. Bank CV (on geometric context space)\n", + " if B >= 10:\n", + " ctx_n = F.normalize(geo_context, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = ctx_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " bank_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"bank_cv\"] = bank_cv\n", + " else:\n", + " aux[\"bank_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # 7. Embedding-space CV (should match consensus target)\n", + " if B >= 10:\n", + " emb_n = F.normalize(emb, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = emb_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " emb_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"emb_cv\"] = emb_cv\n", + " else:\n", + " aux[\"emb_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # Diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"anchor_mean_cos\"] = anchor_cos.mean().item()\n", + " if cross_features.shape[1] > 0:\n", + " aux[\"cross_expert_cos\"] = cross_features.mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux):\n", + " \"\"\"\n", + " All targets come from measured consensus statistics.\n", + " No arbitrary constants.\n", + " \"\"\"\n", + " loss = (\n", + " 1.0 * aux[\"expert_agreement\"] + # experts should agree\n", + " 1.0 * aux[\"rotation_ortho\"] + # rotations stay orthogonal\n", + " 0.5 * aux[\"anchor_spread\"] + # anchors cover the sphere\n", + " 0.1 * aux[\"anchor_entropy\"] + # sharp anchor assignments\n", + " 0.3 * aux[\"cross_expert_var\"] + # stable cross-expert structure\n", + " 0.3 * (aux[\"bank_cv\"] - self.target_cv).abs() + # bank CV → consensus CV\n", + " 0.3 * (aux[\"emb_cv\"] - self.target_cv).abs() # verify emb CV stays at consensus\n", + " )\n", + " return loss\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.12, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def measure_consensus_stats(consensus_embs, n_check=2000):\n", + " \"\"\"Measure exact geometric statistics of the consensus manifold.\"\"\"\n", + " embs = consensus_embs[:n_check].float()\n", + " # CV\n", + " cv = cv_metric(embs.to(DEVICE))\n", + " # Pairwise cosine\n", + " sim = embs @ embs.T\n", + " mask = ~torch.eye(embs.shape[0], dtype=torch.bool)\n", + " pairwise = sim[mask]\n", + " mean_cos = pairwise.mean().item()\n", + " # Spectral\n", + " centered = embs - embs.mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered)\n", + " S_norm = (S / (S.sum() + 1e-8)).tolist()[:50]\n", + " # Eff dim\n", + " eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " return {\n", + " \"cv\": cv,\n", + " \"mean_cos\": mean_cos,\n", + " \"spectral\": S_norm,\n", + " \"eff_dim\": eff_dim,\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION + ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " N_SAMPLES = 20000\n", + " MAX_LEN = 128\n", + " BATCH = 256\n", + "\n", + " # ── Phase 0: Extract ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXTRACTION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_SAMPLES:\n", + " break\n", + " print(f\" Captions: {len(captions):,}\")\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " print(f\"\\n Extracting: {short}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + " embeds[short] = torch.cat(all_emb)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 0b: Align + Consensus + Measure ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0b: PROCRUSTES ALIGNMENT + CONSENSUS STATISTICS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref = \"bert\"\n", + " names = [s for _, s, _ in EXPERTS]\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], embeds[ref])\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " print(f\" {name:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}\")\n", + "\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " for name in names:\n", + " cos = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {cos:.4f}\")\n", + "\n", + " # Measure EXACT consensus statistics — these become the bank's targets\n", + " print(f\"\\n Measuring consensus statistics...\")\n", + " consensus_stats = measure_consensus_stats(consensus)\n", + " print(f\" CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Mean cos: {consensus_stats['mean_cos']:.4f}\")\n", + " print(f\" Eff dim: {consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Spectral: [{', '.join(f'{s:.4f}' for s in consensus_stats['spectral'][:5])}...]\")\n", + "\n", + " del embeds, aligned\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 1: Train Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: TRAIN STUDENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " tokens = tokenizer(captions, max_length=MAX_LEN, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N_SAMPLES - 2000\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " student = MiniStudent(\n", + " vocab_size=tokenizer.vocab_size, max_len=MAX_LEN,\n", + " d_model=256, n_heads=4, n_layers=4, d_ff=1024,\n", + " output_dim=768, dropout=0.1, pad_token_id=tokenizer.pad_token_id\n", + " ).to(DEVICE)\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Student: {n_params:,} params\")\n", + " print(f\" CV target: {consensus_stats['cv']:.4f}\")\n", + "\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=3e-4, weight_decay=0.01)\n", + "\n", + " for epoch in range(5):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " t_loss, t_acc, t_cos, n = 0, 0, 0, 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_stats[\"cv\"])\n", + " loss = l_nce + l_mse + 0.1 * l_cv\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " t_loss += loss.item(); t_acc += acc; t_cos += cos; n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + " student.eval()\n", + " with torch.no_grad():\n", + " v_emb = student(val_ids, val_mask)\n", + " _, v_acc = infonce(v_emb[:1000], val_targets[:1000])\n", + " v_cos = F.cosine_similarity(v_emb, val_targets, dim=-1).mean().item()\n", + " v_cv = cv_metric(v_emb[:1000])\n", + " print(f\" E{epoch+1}: {elapsed:.0f}s loss={t_loss/d:.4f} \"\n", + " f\"t_acc={t_acc/d:.3f} t_cos={t_cos/d:.3f} \"\n", + " f\"v_acc={v_acc:.3f} v_cos={v_cos:.3f} v_cv={v_cv:.3f}\")\n", + "\n", + " torch.save(student.state_dict(), \"mini_student.pt\")\n", + " print(f\"\\n Student saved. v_cos={v_cos:.3f}, v_cv={v_cv:.3f}\")\n", + "\n", + " # ── Phase 2: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: TRAIN ALIGNMENT BANK (student frozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.eval()\n", + " for p in student.parameters():\n", + " p.requires_grad = False\n", + "\n", + " print(\" Pre-encoding through frozen student...\")\n", + " with torch.no_grad():\n", + " all_embs = []\n", + " for i in range(0, n_train, 512):\n", + " j = min(i + 512, n_train)\n", + " emb = student(train_ids[i:j], train_mask[i:j])\n", + " all_embs.append(emb)\n", + " student_embs = torch.cat(all_embs)\n", + " val_student_embs = student(val_ids, val_mask)\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=512, d_bank=128\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names,\n", + " consensus[:n_train], consensus_stats)\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + " print(f\" Bank targets: CV={bank.target_cv.item():.4f}, \"\n", + " f\"mean_cos={bank.target_mean_cos.item():.4f}\")\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=1e-3, weight_decay=0.01)\n", + " BANK_EPOCHS = 20\n", + " BANK_BATCH = 256\n", + "\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0,\n", + " \"anchor_spread\": 0, \"bank_cv\": 0, \"emb_cv\": 0,\n", + " \"cross_expert_var\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + " emb = student_embs[idx]\n", + " enriched, aux = bank(emb)\n", + " loss = bank.bank_loss(aux)\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " v_enriched, v_aux = bank(val_student_embs)\n", + " v_loss = bank.bank_loss(v_aux).item()\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f} \"\n", + " f\"agr={stats['expert_agreement']/d:.5f} \"\n", + " f\"ortho={stats['rotation_ortho']/d:.5f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} \"\n", + " f\"b_cv={stats['bank_cv']/d:.4f} \"\n", + " f\"e_cv={stats['emb_cv']/d:.4f} \"\n", + " f\"x_var={stats['cross_expert_var']/d:.5f} \"\n", + " f\"a_max={v_aux['anchor_max_cos']:.3f} \"\n", + " f\"exp={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank.pt\")\n", + "\n", + " # ── Phase 3: Geometric Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " enriched_val, v_aux = bank(val_student_embs)\n", + " original_768 = enriched_val[:, :768]\n", + " geo_context = enriched_val[:, 768:]\n", + "\n", + " passthrough_cos = F.cosine_similarity(\n", + " original_768[:100], val_student_embs[:100], dim=-1).mean().item()\n", + " geo_cv = cv_metric(F.normalize(geo_context[:1000], dim=-1))\n", + " S = torch.linalg.svdvals(\n", + " geo_context[:1000].float() - geo_context[:1000].float().mean(0))\n", + " geo_eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " # Verify consensus stats are preserved\n", + " emb_cv = cv_metric(val_student_embs[:1000])\n", + "\n", + " print(f\" Passthrough: {passthrough_cos:.6f} (target: 1.000)\")\n", + " print(f\" Emb CV: {emb_cv:.4f} (consensus: {consensus_stats['cv']:.4f})\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo eff_dim: {geo_eff_dim:.1f} / {bank.d_bank}\")\n", + " print(f\" Expert cos: {v_aux['expert_cos_mean']:.3f} ± {v_aux['expert_cos_std']:.3f}\")\n", + " print(f\" Anchor max cos: {v_aux['anchor_max_cos']:.3f}\")\n", + " if \"cross_expert_cos\" in v_aux:\n", + " print(f\" Cross-expert: {v_aux['cross_expert_cos']:.3f}\")\n", + "\n", + " # ── Phase 4: Classifier Stability Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " with torch.no_grad():\n", + " embs = val_student_embs[:1000]\n", + " sim = embs @ embs.T\n", + " sim.fill_diagonal_(-1)\n", + " n_pairs = 3000\n", + " idx_a = torch.randint(0, 1000, (n_pairs,))\n", + " idx_b = torch.randint(0, 1000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1\n", + " labels[pair_cos <= t1] = 2\n", + " enriched_a, _ = bank(embs[idx_a])\n", + " enriched_b, _ = bank(embs[idx_b])\n", + "\n", + " for mode in [\"with_bank\", \"without_bank\"]:\n", + " if mode == \"with_bank\":\n", + " feat_dim = (768 + 128) * 2\n", + " features = torch.cat([enriched_a, enriched_b], dim=-1)\n", + " else:\n", + " feat_dim = 768 * 2\n", + " features = torch.cat([embs[idx_a], embs[idx_b]], dim=-1)\n", + "\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, 256), nn.GELU(), nn.LayerNorm(256),\n", + " nn.Linear(256, 3)\n", + " ).to(DEVICE)\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " n_clf_train = 2400\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + " for e in range(30):\n", + " clf.train()\n", + " logits = clf(train_f)\n", + " loss = F.cross_entropy(logits, train_l)\n", + " loss.backward(); clf_opt.step(); clf_opt.zero_grad()\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " v_acc = (clf(val_f).argmax(-1) == val_l).float().mean().item()\n", + " t_acc = (clf(train_f).argmax(-1) == train_l).float().mean().item()\n", + " print(f\" {mode:15s}: train={t_acc:.3f} val={v_acc:.3f} gap={t_acc-v_acc:.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Consensus CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Consensus eff_dim:{consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Student v_cos: {v_cos:.3f}\")\n", + " print(f\" Student v_cv: {v_cv:.3f}\")\n", + " print(f\" Bank params: {bank_params:,}\")\n", + " print(f\" Bank geo_eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Bank geo_cv: {geo_cv:.4f}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "fb248fb346564862b9c633d195d88d2a", + "68cc85aa9433422d925937b245bad649", + "ed527cdca7de4f569ae2e879ae2c5cc4", + "3926e9cb649844cbaed16d16af96490d", + "19dff49386d24a4cb9940d1c41b6d87c", + "75912febbe024523ab196ebee515abe3", + "a492e1fac92342dab64eac36dd2694ea", + "e00e46d477814ff3a15d3a0549cadcb4", + "2665800f69444ac5b9f7ba2d70eab01d", + "93257a415826429f9de685655370237e", + "c01652b8718e4a80b8c92c82a1fc9f49", + "5c2fe3ce505c425b8a5133636fa3e59a", + "8eaf6eedc46144ea828b8cf1bf3c177a", + "9b4aa70316fd4b30acc0a957b2e024dd", + "dd4163db08b94b88bbdb6f2e914d91a4", + "17938cc8e1184b3ca9db7c72fe5c40fe", + "9549dcb9f1604714a906e581101174fc", + "493fc58cc1084e00b4aef9e1a4ce7687", + "73fdc6a1594c47de86c5124e192bc43f", + "6fbf05deedf94711a079726a73811f52", + "3f36a310c43c4012a2a0c19294fe0e13", + "11b3b34f9ffc4e879152200104225df1" + ] + }, + "id": "KwYHvOk_0FeQ", + "outputId": "07319bb1-c7db-4aea-a173-8b54525c6e01" + }, + "execution_count": 1, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "RAPID PROTOTYPE v2: Differentiation-Centered Bank\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "PHASE 0: EXTRACTION\n", + "=================================================================\n", + " Captions: 20,000\n", + "\n", + " Extracting: bert...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 0:\n", + " aux[\"cross_expert_var\"] = cross_features.var(dim=0).mean()\n", + " else:\n", + " aux[\"cross_expert_var\"] = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # 6. Bank CV (on geometric context space)\n", + " if B >= 10:\n", + " ctx_n = F.normalize(geo_context, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = ctx_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " bank_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"bank_cv\"] = bank_cv\n", + " else:\n", + " aux[\"bank_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # 7. Embedding-space CV (should match consensus target)\n", + " if B >= 10:\n", + " emb_n = F.normalize(emb, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = emb_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " emb_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"emb_cv\"] = emb_cv\n", + " else:\n", + " aux[\"emb_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # Diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"anchor_mean_cos\"] = anchor_cos.mean().item()\n", + " if cross_features.shape[1] > 0:\n", + " aux[\"cross_expert_cos\"] = cross_features.mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux):\n", + " \"\"\"\n", + " All targets come from measured consensus statistics.\n", + " No arbitrary constants.\n", + " \"\"\"\n", + " loss = (\n", + " 1.0 * aux[\"expert_agreement\"] + # experts should agree\n", + " 1.0 * aux[\"rotation_ortho\"] + # rotations stay orthogonal\n", + " 0.5 * aux[\"anchor_spread\"] + # anchors cover the sphere\n", + " 0.1 * aux[\"anchor_entropy\"] + # sharp anchor assignments\n", + " 0.3 * aux[\"cross_expert_var\"] + # stable cross-expert structure\n", + " 0.3 * (aux[\"bank_cv\"] - self.target_cv).abs() + # bank CV → consensus CV\n", + " 0.3 * (aux[\"emb_cv\"] - self.target_cv).abs() # verify emb CV stays at consensus\n", + " )\n", + " return loss\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.12, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def measure_consensus_stats(consensus_embs, n_check=2000):\n", + " \"\"\"Measure exact geometric statistics of the consensus manifold.\"\"\"\n", + " embs = consensus_embs[:n_check].float()\n", + " # CV\n", + " cv = cv_metric(embs.to(DEVICE))\n", + " # Pairwise cosine\n", + " sim = embs @ embs.T\n", + " mask = ~torch.eye(embs.shape[0], dtype=torch.bool)\n", + " pairwise = sim[mask]\n", + " mean_cos = pairwise.mean().item()\n", + " # Spectral\n", + " centered = embs - embs.mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered)\n", + " S_norm = (S / (S.sum() + 1e-8)).tolist()[:50]\n", + " # Eff dim\n", + " eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " return {\n", + " \"cv\": cv,\n", + " \"mean_cos\": mean_cos,\n", + " \"spectral\": S_norm,\n", + " \"eff_dim\": eff_dim,\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION + ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " N_SAMPLES = 20000\n", + " MAX_LEN = 128\n", + " BATCH = 256\n", + "\n", + " # ── Phase 0: Extract ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXTRACTION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_SAMPLES:\n", + " break\n", + " print(f\" Captions: {len(captions):,}\")\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " print(f\"\\n Extracting: {short}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + " embeds[short] = torch.cat(all_emb)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 0b: Align + Consensus + Measure ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0b: PROCRUSTES ALIGNMENT + CONSENSUS STATISTICS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref = \"bert\"\n", + " names = [s for _, s, _ in EXPERTS]\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], embeds[ref])\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " print(f\" {name:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}\")\n", + "\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " for name in names:\n", + " cos = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {cos:.4f}\")\n", + "\n", + " # Measure EXACT consensus statistics — these become the bank's targets\n", + " print(f\"\\n Measuring consensus statistics...\")\n", + " consensus_stats = measure_consensus_stats(consensus)\n", + " print(f\" CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Mean cos: {consensus_stats['mean_cos']:.4f}\")\n", + " print(f\" Eff dim: {consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Spectral: [{', '.join(f'{s:.4f}' for s in consensus_stats['spectral'][:5])}...]\")\n", + "\n", + " del embeds, aligned\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 1: Train Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: TRAIN STUDENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " tokens = tokenizer(captions, max_length=MAX_LEN, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N_SAMPLES - 2000\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " student = MiniStudent(\n", + " vocab_size=tokenizer.vocab_size, max_len=MAX_LEN,\n", + " d_model=256, n_heads=4, n_layers=4, d_ff=1024,\n", + " output_dim=768, dropout=0.1, pad_token_id=tokenizer.pad_token_id\n", + " ).to(DEVICE)\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Student: {n_params:,} params\")\n", + " print(f\" CV target: {consensus_stats['cv']:.4f}\")\n", + "\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=3e-4, weight_decay=0.01)\n", + "\n", + " for epoch in range(5):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " t_loss, t_acc, t_cos, n = 0, 0, 0, 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_stats[\"cv\"])\n", + " loss = l_nce + l_mse + 0.1 * l_cv\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " t_loss += loss.item(); t_acc += acc; t_cos += cos; n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + " student.eval()\n", + " with torch.no_grad():\n", + " v_emb = student(val_ids, val_mask)\n", + " _, v_acc = infonce(v_emb[:1000], val_targets[:1000])\n", + " v_cos = F.cosine_similarity(v_emb, val_targets, dim=-1).mean().item()\n", + " v_cv = cv_metric(v_emb[:1000])\n", + " print(f\" E{epoch+1}: {elapsed:.0f}s loss={t_loss/d:.4f} \"\n", + " f\"t_acc={t_acc/d:.3f} t_cos={t_cos/d:.3f} \"\n", + " f\"v_acc={v_acc:.3f} v_cos={v_cos:.3f} v_cv={v_cv:.3f}\")\n", + "\n", + " torch.save(student.state_dict(), \"mini_student.pt\")\n", + " print(f\"\\n Student saved. v_cos={v_cos:.3f}, v_cv={v_cv:.3f}\")\n", + "\n", + " # ── Phase 2: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: TRAIN ALIGNMENT BANK (student frozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.eval()\n", + " for p in student.parameters():\n", + " p.requires_grad = False\n", + "\n", + " print(\" Pre-encoding through frozen student...\")\n", + " with torch.no_grad():\n", + " all_embs = []\n", + " for i in range(0, n_train, 512):\n", + " j = min(i + 512, n_train)\n", + " emb = student(train_ids[i:j], train_mask[i:j])\n", + " all_embs.append(emb)\n", + " student_embs = torch.cat(all_embs)\n", + " val_student_embs = student(val_ids, val_mask)\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=512, d_bank=128\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names,\n", + " consensus[:n_train], consensus_stats)\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + " print(f\" Bank targets: CV={bank.target_cv.item():.4f}, \"\n", + " f\"mean_cos={bank.target_mean_cos.item():.4f}\")\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=1e-3, weight_decay=0.01)\n", + " BANK_EPOCHS = 20\n", + " BANK_BATCH = 256\n", + "\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0,\n", + " \"anchor_spread\": 0, \"bank_cv\": 0, \"emb_cv\": 0,\n", + " \"cross_expert_var\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + " emb = student_embs[idx]\n", + " enriched, aux = bank(emb)\n", + " loss = bank.bank_loss(aux)\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " v_enriched, v_aux = bank(val_student_embs)\n", + " v_loss = bank.bank_loss(v_aux).item()\n", + "\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f} \"\n", + " f\"agr={stats['expert_agreement']/d:.5f} \"\n", + " f\"ortho={stats['rotation_ortho']/d:.5f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} \"\n", + " f\"b_cv={stats['bank_cv']/d:.4f} \"\n", + " f\"e_cv={stats['emb_cv']/d:.4f} \"\n", + " f\"x_var={stats['cross_expert_var']/d:.5f} \"\n", + " f\"a_max={v_aux['anchor_max_cos']:.3f} \"\n", + " f\"exp={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank.pt\")\n", + "\n", + " # ── Phase 3: Geometric Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " enriched_val, v_aux = bank(val_student_embs)\n", + " original_768 = enriched_val[:, :768]\n", + " geo_context = enriched_val[:, 768:]\n", + "\n", + " passthrough_cos = F.cosine_similarity(\n", + " original_768[:100], val_student_embs[:100], dim=-1).mean().item()\n", + " geo_cv = cv_metric(F.normalize(geo_context[:1000], dim=-1))\n", + " S = torch.linalg.svdvals(\n", + " geo_context[:1000].float() - geo_context[:1000].float().mean(0))\n", + " geo_eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " # Verify consensus stats are preserved\n", + " emb_cv = cv_metric(val_student_embs[:1000])\n", + "\n", + " print(f\" Passthrough: {passthrough_cos:.6f} (target: 1.000)\")\n", + " print(f\" Emb CV: {emb_cv:.4f} (consensus: {consensus_stats['cv']:.4f})\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo eff_dim: {geo_eff_dim:.1f} / {bank.d_bank}\")\n", + " print(f\" Expert cos: {v_aux['expert_cos_mean']:.3f} ± {v_aux['expert_cos_std']:.3f}\")\n", + " print(f\" Anchor max cos: {v_aux['anchor_max_cos']:.3f}\")\n", + " if \"cross_expert_cos\" in v_aux:\n", + " print(f\" Cross-expert: {v_aux['cross_expert_cos']:.3f}\")\n", + "\n", + " # ── Phase 4: Classifier Stability Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " with torch.no_grad():\n", + " embs = val_student_embs[:1000]\n", + " sim = embs @ embs.T\n", + " sim.fill_diagonal_(-1)\n", + " n_pairs = 3000\n", + " idx_a = torch.randint(0, 1000, (n_pairs,))\n", + " idx_b = torch.randint(0, 1000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1\n", + " labels[pair_cos <= t1] = 2\n", + " enriched_a, _ = bank(embs[idx_a])\n", + " enriched_b, _ = bank(embs[idx_b])\n", + "\n", + " for mode in [\"with_bank\", \"without_bank\"]:\n", + " if mode == \"with_bank\":\n", + " feat_dim = (768 + 128) * 2\n", + " features = torch.cat([enriched_a, enriched_b], dim=-1)\n", + " else:\n", + " feat_dim = 768 * 2\n", + " features = torch.cat([embs[idx_a], embs[idx_b]], dim=-1)\n", + "\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, 256), nn.GELU(), nn.LayerNorm(256),\n", + " nn.Linear(256, 3)\n", + " ).to(DEVICE)\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " n_clf_train = 2400\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + " for e in range(30):\n", + " clf.train()\n", + " logits = clf(train_f)\n", + " loss = F.cross_entropy(logits, train_l)\n", + " loss.backward(); clf_opt.step(); clf_opt.zero_grad()\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " v_acc = (clf(val_f).argmax(-1) == val_l).float().mean().item()\n", + " t_acc = (clf(train_f).argmax(-1) == train_l).float().mean().item()\n", + " print(f\" {mode:15s}: train={t_acc:.3f} val={v_acc:.3f} gap={t_acc-v_acc:.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Consensus CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Consensus eff_dim:{consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Student v_cos: {v_cos:.3f}\")\n", + " print(f\" Student v_cv: {v_cv:.3f}\")\n", + " print(f\" Bank params: {bank_params:,}\")\n", + " print(f\" Bank geo_eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Bank geo_cv: {geo_cv:.4f}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "96e6b188456a440491abf1b9f97e0f1b", + "2bbf292baba24d83bc6969f17945ebdf", + "bfbcf1c77d2e4c6dab61ecdb8b32f45b", + "c9380e44797d4631bacdce608af8501d", + "e4d3027a8ed1412693b1e8dd601c06ab", + "f63ba9edd18d483795ebaa2ac160dbed", + "2d01e0821f704e09945bcd17419c11e2", + "e89e69591d21429fa8007b0fa576d61c", + "5cdba3bf3b5f4d259ebd63ecab967fc2", + "169c779b68ca4da6a16eb88b87e0cf62", + "c4bdcb3665724031a1e7f1f06cfd496e", + "1be48dcc1d5a44af98283079bf2e251c", + "d00dd42d3bd5429d86f33edaa9a85b8b", + "c219a9706b674b7aa89c4d5bc11e6df3", + "1f772e7517fe4ef5bbcc657a384d37e8", + "eadbb240982c4fb6aa19bf61ae99c672", + "b06994fee46d41d7930ce03d439a1ae9", + "0c37b804865244e9b4db2aac3e277a6e", + "3284c3d7ea4342dbae5387da632d73f4", + "3bbc747f5ccf4acab6ae9c38506fef28", + "4af6797d8e78453982f35694c60785ef", + "a1f0417b8f464996b1c93dc219c3d22f" + ] + }, + "id": "YmyhkFVO1nx0", + "outputId": "07471f82-6a7d-46c9-a95b-c3811b59273d" + }, + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "RAPID PROTOTYPE v2: Differentiation-Centered Bank\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "PHASE 0: EXTRACTION\n", + "=================================================================\n", + " Captions: 20,000\n", + "\n", + " Extracting: bert...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 0:\n", + " aux[\"cross_expert_var\"] = cross_features.var(dim=0).mean()\n", + " else:\n", + " aux[\"cross_expert_var\"] = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # 6. Disagreement preservation\n", + " # The distribution of disagreement should stay at the measured target\n", + " batch_cross_mean = cross_features.mean() if cross_features.shape[1] > 0 else torch.tensor(0.0, device=emb.device)\n", + " batch_cross_std = cross_features.std() if cross_features.shape[1] > 0 else torch.tensor(0.0, device=emb.device)\n", + " batch_disagree_ratio = disagreement_ratio.mean()\n", + " aux[\"disagree_preserve\"] = (\n", + " (batch_cross_mean - self.target_cross_cos_mean).pow(2) +\n", + " (batch_cross_std - self.target_cross_cos_std).pow(2) +\n", + " (batch_disagree_ratio - self.target_disagreement_ratio).pow(2)\n", + " )\n", + "\n", + " # 7. Bank CV\n", + " if B >= 10:\n", + " ctx_n = F.normalize(geo_context, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = ctx_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " bank_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"bank_cv\"] = bank_cv\n", + " else:\n", + " aux[\"bank_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # 8. Emb CV\n", + " if B >= 10:\n", + " emb_n = F.normalize(emb, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = emb_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " emb_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"emb_cv\"] = emb_cv\n", + " else:\n", + " aux[\"emb_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # Diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"anchor_mean_cos\"] = anchor_cos.mean().item()\n", + " if cross_features.shape[1] > 0:\n", + " aux[\"cross_expert_cos\"] = cross_features.mean().item()\n", + " aux[\"cross_expert_cos_std\"] = cross_features.std().item()\n", + " aux[\"disagreement_ratio\"] = disagreement_ratio.mean().item()\n", + " aux[\"norm_ratio_spread\"] = norm_ratio.std(dim=-1).mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux):\n", + " \"\"\"All targets from measured consensus. Preserves disagreement structure.\"\"\"\n", + " loss = (\n", + " 1.0 * aux[\"expert_agreement\"] +\n", + " 1.0 * aux[\"rotation_ortho\"] +\n", + " 0.5 * aux[\"anchor_spread\"] +\n", + " 0.1 * aux[\"anchor_entropy\"] +\n", + " 0.3 * aux[\"cross_expert_var\"] +\n", + " 0.3 * (aux[\"bank_cv\"] - self.target_cv).abs() +\n", + " 0.3 * (aux[\"emb_cv\"] - self.target_cv).abs() +\n", + " 0.5 * aux[\"disagree_preserve\"] # preserve the disagreement distribution\n", + " )\n", + " return loss\n", + "\n", + " @torch.no_grad()\n", + " def calibrate_disagreement(self, embeddings):\n", + " \"\"\"\n", + " Measure the initial disagreement structure and store as targets.\n", + " Call ONCE after init, before training.\n", + " \"\"\"\n", + " _, aux = self.forward(embeddings)\n", + " if \"cross_expert_cos\" in aux:\n", + " self.target_cross_cos_mean.fill_(aux[\"cross_expert_cos\"])\n", + " if \"cross_expert_cos_std\" in aux:\n", + " self.target_cross_cos_std.fill_(aux[\"cross_expert_cos_std\"])\n", + " self.target_disagreement_ratio.fill_(aux[\"disagreement_ratio\"])\n", + " print(f\" Calibrated disagreement:\")\n", + " print(f\" cross_cos: {self.target_cross_cos_mean.item():.4f} ± {self.target_cross_cos_std.item():.4f}\")\n", + " print(f\" disagree_ratio: {self.target_disagreement_ratio.item():.6f}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.12, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def measure_consensus_stats(consensus_embs, n_check=2000):\n", + " \"\"\"Measure exact geometric statistics of the consensus manifold.\"\"\"\n", + " embs = consensus_embs[:n_check].float()\n", + " # CV\n", + " cv = cv_metric(embs.to(DEVICE))\n", + " # Pairwise cosine\n", + " sim = embs @ embs.T\n", + " mask = ~torch.eye(embs.shape[0], dtype=torch.bool)\n", + " pairwise = sim[mask]\n", + " mean_cos = pairwise.mean().item()\n", + " # Spectral\n", + " centered = embs - embs.mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered)\n", + " S_norm = (S / (S.sum() + 1e-8)).tolist()[:50]\n", + " # Eff dim\n", + " eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " return {\n", + " \"cv\": cv,\n", + " \"mean_cos\": mean_cos,\n", + " \"spectral\": S_norm,\n", + " \"eff_dim\": eff_dim,\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION + ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " N_SAMPLES = 20000\n", + " MAX_LEN = 128\n", + " BATCH = 256\n", + "\n", + " # ── Phase 0: Extract ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXTRACTION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_SAMPLES:\n", + " break\n", + " print(f\" Captions: {len(captions):,}\")\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " print(f\"\\n Extracting: {short}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + " embeds[short] = torch.cat(all_emb)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 0b: Align + Consensus + Measure ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0b: PROCRUSTES ALIGNMENT + CONSENSUS STATISTICS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ref = \"bert\"\n", + " names = [s for _, s, _ in EXPERTS]\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], embeds[ref])\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " print(f\" {name:10s}: cos {info['cos_before']:.4f} → {info['cos_after']:.4f}\")\n", + "\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " for name in names:\n", + " cos = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {cos:.4f}\")\n", + "\n", + " # Measure EXACT consensus statistics — these become the bank's targets\n", + " print(f\"\\n Measuring consensus statistics...\")\n", + " consensus_stats = measure_consensus_stats(consensus)\n", + " print(f\" CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Mean cos: {consensus_stats['mean_cos']:.4f}\")\n", + " print(f\" Eff dim: {consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Spectral: [{', '.join(f'{s:.4f}' for s in consensus_stats['spectral'][:5])}...]\")\n", + "\n", + " del embeds, aligned\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 1: Train Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: TRAIN STUDENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " tokens = tokenizer(captions, max_length=MAX_LEN, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N_SAMPLES - 2000\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " student = MiniStudent(\n", + " vocab_size=tokenizer.vocab_size, max_len=MAX_LEN,\n", + " d_model=256, n_heads=4, n_layers=4, d_ff=1024,\n", + " output_dim=768, dropout=0.1, pad_token_id=tokenizer.pad_token_id\n", + " ).to(DEVICE)\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Student: {n_params:,} params\")\n", + " print(f\" CV target: {consensus_stats['cv']:.4f}\")\n", + "\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=3e-4, weight_decay=0.01)\n", + "\n", + " for epoch in range(5):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " t_loss, t_acc, t_cos, n = 0, 0, 0, 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_stats[\"cv\"])\n", + " loss = l_nce + l_mse + 0.1 * l_cv\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " t_loss += loss.item(); t_acc += acc; t_cos += cos; n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + " student.eval()\n", + " with torch.no_grad():\n", + " v_emb = student(val_ids, val_mask)\n", + " _, v_acc = infonce(v_emb[:1000], val_targets[:1000])\n", + " v_cos = F.cosine_similarity(v_emb, val_targets, dim=-1).mean().item()\n", + " v_cv = cv_metric(v_emb[:1000])\n", + " print(f\" E{epoch+1}: {elapsed:.0f}s loss={t_loss/d:.4f} \"\n", + " f\"t_acc={t_acc/d:.3f} t_cos={t_cos/d:.3f} \"\n", + " f\"v_acc={v_acc:.3f} v_cos={v_cos:.3f} v_cv={v_cv:.3f}\")\n", + "\n", + " torch.save(student.state_dict(), \"mini_student.pt\")\n", + " print(f\"\\n Student saved. v_cos={v_cos:.3f}, v_cv={v_cv:.3f}\")\n", + "\n", + " # ── Phase 2: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: TRAIN ALIGNMENT BANK (student frozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.eval()\n", + " for p in student.parameters():\n", + " p.requires_grad = False\n", + "\n", + " print(\" Pre-encoding through frozen student...\")\n", + " with torch.no_grad():\n", + " all_embs = []\n", + " for i in range(0, n_train, 512):\n", + " j = min(i + 512, n_train)\n", + " emb = student(train_ids[i:j], train_mask[i:j])\n", + " all_embs.append(emb)\n", + " student_embs = torch.cat(all_embs)\n", + " val_student_embs = student(val_ids, val_mask)\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=512, d_bank=128\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names,\n", + " consensus[:n_train], consensus_stats)\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + " print(f\" Bank targets: CV={bank.target_cv.item():.4f}, \"\n", + " f\"mean_cos={bank.target_mean_cos.item():.4f}\")\n", + "\n", + " # Calibrate disagreement from initial state (before any training)\n", + " bank.calibrate_disagreement(student_embs[:2000])\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=1e-3, weight_decay=0.01)\n", + " BANK_EPOCHS = 20\n", + " BANK_BATCH = 256\n", + "\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0,\n", + " \"anchor_spread\": 0, \"bank_cv\": 0, \"emb_cv\": 0,\n", + " \"cross_expert_var\": 0, \"disagree_preserve\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + " emb = student_embs[idx]\n", + " enriched, aux = bank(emb)\n", + " loss = bank.bank_loss(aux)\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " v_enriched, v_aux = bank(val_student_embs)\n", + " v_loss = bank.bank_loss(v_aux).item()\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f}\")\n", + " print(f\" Geometry: b_cv={stats['bank_cv']/d:.4f} e_cv={stats['emb_cv']/d:.4f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} a_max={v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Experts: cos={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f} \"\n", + " f\"agr={stats['expert_agreement']/d:.6f} ortho={stats['rotation_ortho']/d:.6f}\")\n", + " print(f\" Disagree: x_cos={v_aux.get('cross_expert_cos', 0):.4f}±{v_aux.get('cross_expert_cos_std', 0):.4f} \"\n", + " f\"ratio={v_aux['disagreement_ratio']:.6f} \"\n", + " f\"preserve={stats['disagree_preserve']/d:.6f} \"\n", + " f\"norms={v_aux['norm_ratio_spread']:.4f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank.pt\")\n", + "\n", + " # ── Phase 3: Geometric Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " enriched_val, v_aux = bank(val_student_embs)\n", + " original_768 = enriched_val[:, :768]\n", + " geo_context = enriched_val[:, 768:]\n", + "\n", + " passthrough_cos = F.cosine_similarity(\n", + " original_768[:100], val_student_embs[:100], dim=-1).mean().item()\n", + " geo_cv = cv_metric(F.normalize(geo_context[:1000], dim=-1))\n", + " S = torch.linalg.svdvals(\n", + " geo_context[:1000].float() - geo_context[:1000].float().mean(0))\n", + " geo_eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " # Verify consensus stats are preserved\n", + " emb_cv = cv_metric(val_student_embs[:1000])\n", + "\n", + " print(f\" Passthrough: {passthrough_cos:.6f} (target: 1.000)\")\n", + " print(f\" Emb CV: {emb_cv:.4f} (consensus: {consensus_stats['cv']:.4f})\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo eff_dim: {geo_eff_dim:.1f} / {bank.d_bank}\")\n", + " print(f\" Expert cos: {v_aux['expert_cos_mean']:.3f} ± {v_aux['expert_cos_std']:.3f}\")\n", + " print(f\" Anchor max cos: {v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Disagreement:\")\n", + " print(f\" Cross-expert: {v_aux.get('cross_expert_cos', 0):.4f} ± {v_aux.get('cross_expert_cos_std', 0):.4f}\")\n", + " print(f\" Ratio: {v_aux['disagreement_ratio']:.6f} (target: {bank.target_disagreement_ratio.item():.6f})\")\n", + " print(f\" Norm spread: {v_aux['norm_ratio_spread']:.4f}\")\n", + "\n", + " # ── Phase 4: Classifier Stability Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " with torch.no_grad():\n", + " embs = val_student_embs[:1000]\n", + " sim = embs @ embs.T\n", + " sim.fill_diagonal_(-1)\n", + " n_pairs = 3000\n", + " idx_a = torch.randint(0, 1000, (n_pairs,))\n", + " idx_b = torch.randint(0, 1000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1\n", + " labels[pair_cos <= t1] = 2\n", + " enriched_a, _ = bank(embs[idx_a])\n", + " enriched_b, _ = bank(embs[idx_b])\n", + "\n", + " for mode in [\"with_bank\", \"without_bank\"]:\n", + " if mode == \"with_bank\":\n", + " feat_dim = (768 + 128) * 2\n", + " features = torch.cat([enriched_a, enriched_b], dim=-1)\n", + " else:\n", + " feat_dim = 768 * 2\n", + " features = torch.cat([embs[idx_a], embs[idx_b]], dim=-1)\n", + "\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, 256), nn.GELU(), nn.LayerNorm(256),\n", + " nn.Linear(256, 3)\n", + " ).to(DEVICE)\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " n_clf_train = 2400\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + " for e in range(30):\n", + " clf.train()\n", + " logits = clf(train_f)\n", + " loss = F.cross_entropy(logits, train_l)\n", + " loss.backward(); clf_opt.step(); clf_opt.zero_grad()\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " v_acc = (clf(val_f).argmax(-1) == val_l).float().mean().item()\n", + " t_acc = (clf(train_f).argmax(-1) == train_l).float().mean().item()\n", + " print(f\" {mode:15s}: train={t_acc:.3f} val={v_acc:.3f} gap={t_acc-v_acc:.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Consensus CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Consensus eff_dim:{consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Student v_cos: {v_cos:.3f}\")\n", + " print(f\" Student v_cv: {v_cv:.3f}\")\n", + " print(f\" Bank params: {bank_params:,}\")\n", + " print(f\" Bank geo_eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Bank geo_cv: {geo_cv:.4f}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "3fc59d54447c48cf95c06fa0b7c44f81", + "97a9f9336df3492187221ef28aa2be6c", + "44a2bc1c6841464ab8a914aa7907f0e9", + "2b914a27903c485faa75b8f755bffe6e", + "4caeb796ecc642148f1921227a0ffa76", + "e892f44899514757af1188a6e80b7382", + "5b31210bff804a90802064e36ed51938", + "a46ef317e76345dda61ce0ce5eb61181", + "650619cadc124853993e836b3fae340e", + "df54dc7f2cfb496d923f85aa9aa719f7", + "9258ee70426b4117ae3ae148f6332581", + "9faa8b9c10da45b6b3088d2246dab05b", + "926ffa40ae224a61927901a0b7524251", + "27802385da91475ab1b765fee5519d87", + "7265f91c04194da3a4abf7c7b8bedff2", + "c2c5c0a854ee477680bdffb98b0d8a89", + "75a32209b14e466ab66498a2a7894b28", + "ae3a834c74d34f7e842764df0ea2ae44", + "41a228039e1843dc92e771715040d525", + "b2c514fc44ba465bb6d77c6097f3ddf7", + "760d3b2292ba4f67a9927387024169d7", + "7da5505ebc93451a81f4c5b7b8a72ee1" + ] + }, + "id": "xXlp5Wh05pmM", + "outputId": "2ea3d4a5-de09-4995-aa52-cc93c65306fd" + }, + "execution_count": 4, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "RAPID PROTOTYPE v2: Differentiation-Centered Bank\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "PHASE 0: EXTRACTION\n", + "=================================================================\n", + " Captions: 20,000\n", + "\n", + " Extracting: bert...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 0:\n", + " aux[\"cross_expert_var\"] = cross_features.var(dim=0).mean()\n", + " else:\n", + " aux[\"cross_expert_var\"] = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # 6. Disagreement preservation\n", + " # The distribution of disagreement should stay at the measured target\n", + " batch_cross_mean = cross_features.mean() if cross_features.shape[1] > 0 else torch.tensor(0.0, device=emb.device)\n", + " batch_cross_std = cross_features.std() if cross_features.shape[1] > 0 else torch.tensor(0.0, device=emb.device)\n", + " batch_disagree_ratio = disagreement_ratio.mean()\n", + " aux[\"disagree_preserve\"] = (\n", + " (batch_cross_mean - self.target_cross_cos_mean).pow(2) +\n", + " (batch_cross_std - self.target_cross_cos_std).pow(2) +\n", + " (batch_disagree_ratio - self.target_disagreement_ratio).pow(2)\n", + " )\n", + "\n", + " # 7. Bank CV\n", + " if B >= 10:\n", + " ctx_n = F.normalize(geo_context, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = ctx_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " bank_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"bank_cv\"] = bank_cv\n", + " else:\n", + " aux[\"bank_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # 8. Emb CV\n", + " if B >= 10:\n", + " emb_n = F.normalize(emb, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = emb_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " emb_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"emb_cv\"] = emb_cv\n", + " else:\n", + " aux[\"emb_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # Diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"anchor_mean_cos\"] = anchor_cos.mean().item()\n", + " if cross_features.shape[1] > 0:\n", + " aux[\"cross_expert_cos\"] = cross_features.mean().item()\n", + " aux[\"cross_expert_cos_std\"] = cross_features.std().item()\n", + " aux[\"disagreement_ratio\"] = disagreement_ratio.mean().item()\n", + " aux[\"norm_ratio_spread\"] = norm_ratio.std(dim=-1).mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux):\n", + " \"\"\"All targets from measured consensus. Preserves disagreement structure.\"\"\"\n", + " loss = (\n", + " 1.0 * aux[\"expert_agreement\"] +\n", + " 1.0 * aux[\"rotation_ortho\"] +\n", + " 0.5 * aux[\"anchor_spread\"] +\n", + " 0.1 * aux[\"anchor_entropy\"] +\n", + " 0.3 * aux[\"cross_expert_var\"] +\n", + " 0.3 * (aux[\"bank_cv\"] - self.target_cv).abs() +\n", + " 0.3 * (aux[\"emb_cv\"] - self.target_cv).abs() +\n", + " 0.5 * aux[\"disagree_preserve\"] # preserve the disagreement distribution\n", + " )\n", + " return loss\n", + "\n", + " @torch.no_grad()\n", + " def calibrate_disagreement(self, embeddings):\n", + " \"\"\"\n", + " Measure the initial disagreement structure and store as targets.\n", + " Call ONCE after init, before training.\n", + " \"\"\"\n", + " _, aux = self.forward(embeddings)\n", + " if \"cross_expert_cos\" in aux:\n", + " self.target_cross_cos_mean.fill_(aux[\"cross_expert_cos\"])\n", + " if \"cross_expert_cos_std\" in aux:\n", + " self.target_cross_cos_std.fill_(aux[\"cross_expert_cos_std\"])\n", + " self.target_disagreement_ratio.fill_(aux[\"disagreement_ratio\"])\n", + " print(f\" Calibrated disagreement:\")\n", + " print(f\" cross_cos: {self.target_cross_cos_mean.item():.4f} ± {self.target_cross_cos_std.item():.4f}\")\n", + " print(f\" disagree_ratio: {self.target_disagreement_ratio.item():.6f}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.12, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def measure_consensus_stats(consensus_embs, n_check=2000):\n", + " \"\"\"Measure exact geometric statistics of the consensus manifold.\"\"\"\n", + " embs = consensus_embs[:n_check].float()\n", + " # CV\n", + " cv = cv_metric(embs.to(DEVICE))\n", + " # Pairwise cosine\n", + " sim = embs @ embs.T\n", + " mask = ~torch.eye(embs.shape[0], dtype=torch.bool)\n", + " pairwise = sim[mask]\n", + " mean_cos = pairwise.mean().item()\n", + " # Spectral\n", + " centered = embs - embs.mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered)\n", + " S_norm = (S / (S.sum() + 1e-8)).tolist()[:50]\n", + " # Eff dim\n", + " eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " return {\n", + " \"cv\": cv,\n", + " \"mean_cos\": mean_cos,\n", + " \"spectral\": S_norm,\n", + " \"eff_dim\": eff_dim,\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION + ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " N_SAMPLES = 20000\n", + " MAX_LEN = 128\n", + " BATCH = 256\n", + "\n", + " # ── Phase 0: Extract ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXTRACTION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_SAMPLES:\n", + " break\n", + " print(f\" Captions: {len(captions):,}\")\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " print(f\"\\n Extracting: {short}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + " embeds[short] = torch.cat(all_emb)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 0b: Align + Consensus + Measure ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0b: GENERALIZED PROCRUSTES ALIGNMENT (no reference bias)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " names = [s for _, s, _ in EXPERTS]\n", + "\n", + " # Generalized Procrustes: iteratively align all to their mean\n", + " # No expert is the reference. The centerpoint emerges.\n", + " GPA_ITERS = 10\n", + " current = {name: embeds[name].float() for name in names}\n", + "\n", + " for gpa_iter in range(GPA_ITERS):\n", + " # Compute mean shape\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + "\n", + " # Align each to mean\n", + " new_current = {}\n", + " total_delta = 0.0\n", + " for name in names:\n", + " info = procrustes_align(current[name], mean_shape)\n", + " new_current[name] = apply_align(current[name], info)\n", + " # Measure how much this iteration changed things\n", + " delta = (new_current[name] - current[name]).pow(2).mean().item()\n", + " total_delta += delta\n", + "\n", + " current = new_current\n", + " if gpa_iter == 0 or (gpa_iter + 1) % 3 == 0 or total_delta < 1e-8:\n", + " print(f\" GPA iter {gpa_iter+1}: delta={total_delta:.8f}\")\n", + " if total_delta < 1e-8:\n", + " print(f\" Converged at iteration {gpa_iter+1}\")\n", + " break\n", + "\n", + " # Final alignment: align each expert to the converged mean\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], mean_shape)\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " cos = F.cosine_similarity(\n", + " aligned[name][:2000], mean_shape[:2000], dim=-1).mean().item()\n", + " print(f\" {name:10s}: cos_after={info['cos_after']:.4f} cos_to_mean={cos:.4f}\")\n", + "\n", + " # Consensus = normalized centroid (now equidistant from all experts)\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " for name in names:\n", + " cos = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {cos:.4f}\")\n", + "\n", + " # Verify equidistance\n", + " expert_cos_to_consensus = []\n", + " for name in names:\n", + " c = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " expert_cos_to_consensus.append(c)\n", + " equidist_range = max(expert_cos_to_consensus) - min(expert_cos_to_consensus)\n", + " print(f\" Equidistance range: {equidist_range:.4f} (should be near 0)\")\n", + "\n", + " # Measure EXACT consensus statistics\n", + " print(f\"\\n Measuring consensus statistics...\")\n", + " consensus_stats = measure_consensus_stats(consensus)\n", + " print(f\" CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Mean cos: {consensus_stats['mean_cos']:.4f}\")\n", + " print(f\" Eff dim: {consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Spectral: [{', '.join(f'{s:.4f}' for s in consensus_stats['spectral'][:5])}...]\")\n", + "\n", + " del embeds, aligned\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 1: Train Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: TRAIN STUDENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " tokens = tokenizer(captions, max_length=MAX_LEN, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N_SAMPLES - 2000\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " student = MiniStudent(\n", + " vocab_size=tokenizer.vocab_size, max_len=MAX_LEN,\n", + " d_model=256, n_heads=4, n_layers=4, d_ff=1024,\n", + " output_dim=768, dropout=0.1, pad_token_id=tokenizer.pad_token_id\n", + " ).to(DEVICE)\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Student: {n_params:,} params\")\n", + " print(f\" CV target: {consensus_stats['cv']:.4f}\")\n", + "\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=3e-4, weight_decay=0.01)\n", + "\n", + " for epoch in range(5):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " t_loss, t_acc, t_cos, n = 0, 0, 0, 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_stats[\"cv\"])\n", + " loss = l_nce + l_mse + 0.1 * l_cv\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " t_loss += loss.item(); t_acc += acc; t_cos += cos; n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + " student.eval()\n", + " with torch.no_grad():\n", + " v_emb = student(val_ids, val_mask)\n", + " _, v_acc = infonce(v_emb[:1000], val_targets[:1000])\n", + " v_cos = F.cosine_similarity(v_emb, val_targets, dim=-1).mean().item()\n", + " v_cv = cv_metric(v_emb[:1000])\n", + " print(f\" E{epoch+1}: {elapsed:.0f}s loss={t_loss/d:.4f} \"\n", + " f\"t_acc={t_acc/d:.3f} t_cos={t_cos/d:.3f} \"\n", + " f\"v_acc={v_acc:.3f} v_cos={v_cos:.3f} v_cv={v_cv:.3f}\")\n", + "\n", + " torch.save(student.state_dict(), \"mini_student.pt\")\n", + " print(f\"\\n Student saved. v_cos={v_cos:.3f}, v_cv={v_cv:.3f}\")\n", + "\n", + " # ── Phase 2: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: TRAIN ALIGNMENT BANK (student frozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.eval()\n", + " for p in student.parameters():\n", + " p.requires_grad = False\n", + "\n", + " print(\" Pre-encoding through frozen student...\")\n", + " with torch.no_grad():\n", + " all_embs = []\n", + " for i in range(0, n_train, 512):\n", + " j = min(i + 512, n_train)\n", + " emb = student(train_ids[i:j], train_mask[i:j])\n", + " all_embs.append(emb)\n", + " student_embs = torch.cat(all_embs)\n", + " val_student_embs = student(val_ids, val_mask)\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=512, d_bank=128\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names,\n", + " consensus[:n_train], consensus_stats)\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + " print(f\" Bank targets: CV={bank.target_cv.item():.4f}, \"\n", + " f\"mean_cos={bank.target_mean_cos.item():.4f}\")\n", + "\n", + " # Calibrate disagreement from initial state (before any training)\n", + " bank.calibrate_disagreement(student_embs[:2000])\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=1e-3, weight_decay=0.01)\n", + " BANK_EPOCHS = 20\n", + " BANK_BATCH = 256\n", + "\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0,\n", + " \"anchor_spread\": 0, \"bank_cv\": 0, \"emb_cv\": 0,\n", + " \"cross_expert_var\": 0, \"disagree_preserve\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + " emb = student_embs[idx]\n", + " enriched, aux = bank(emb)\n", + " loss = bank.bank_loss(aux)\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " v_enriched, v_aux = bank(val_student_embs)\n", + " v_loss = bank.bank_loss(v_aux).item()\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f}\")\n", + " print(f\" Geometry: b_cv={stats['bank_cv']/d:.4f} e_cv={stats['emb_cv']/d:.4f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} a_max={v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Experts: cos={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f} \"\n", + " f\"agr={stats['expert_agreement']/d:.6f} ortho={stats['rotation_ortho']/d:.6f}\")\n", + " print(f\" Disagree: x_cos={v_aux.get('cross_expert_cos', 0):.4f}±{v_aux.get('cross_expert_cos_std', 0):.4f} \"\n", + " f\"ratio={v_aux['disagreement_ratio']:.6f} \"\n", + " f\"preserve={stats['disagree_preserve']/d:.6f} \"\n", + " f\"norms={v_aux['norm_ratio_spread']:.4f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank.pt\")\n", + "\n", + " # ── Phase 3: Geometric Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " enriched_val, v_aux = bank(val_student_embs)\n", + " original_768 = enriched_val[:, :768]\n", + " geo_context = enriched_val[:, 768:]\n", + "\n", + " passthrough_cos = F.cosine_similarity(\n", + " original_768[:100], val_student_embs[:100], dim=-1).mean().item()\n", + " geo_cv = cv_metric(F.normalize(geo_context[:1000], dim=-1))\n", + " S = torch.linalg.svdvals(\n", + " geo_context[:1000].float() - geo_context[:1000].float().mean(0))\n", + " geo_eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " # Verify consensus stats are preserved\n", + " emb_cv = cv_metric(val_student_embs[:1000])\n", + "\n", + " print(f\" Passthrough: {passthrough_cos:.6f} (target: 1.000)\")\n", + " print(f\" Emb CV: {emb_cv:.4f} (consensus: {consensus_stats['cv']:.4f})\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo eff_dim: {geo_eff_dim:.1f} / {bank.d_bank}\")\n", + " print(f\" Expert cos: {v_aux['expert_cos_mean']:.3f} ± {v_aux['expert_cos_std']:.3f}\")\n", + " print(f\" Anchor max cos: {v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Disagreement:\")\n", + " print(f\" Cross-expert: {v_aux.get('cross_expert_cos', 0):.4f} ± {v_aux.get('cross_expert_cos_std', 0):.4f}\")\n", + " print(f\" Ratio: {v_aux['disagreement_ratio']:.6f} (target: {bank.target_disagreement_ratio.item():.6f})\")\n", + " print(f\" Norm spread: {v_aux['norm_ratio_spread']:.4f}\")\n", + "\n", + " # ── Phase 4: Classifier Stability Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " with torch.no_grad():\n", + " embs = val_student_embs[:1000]\n", + " sim = embs @ embs.T\n", + " sim.fill_diagonal_(-1)\n", + " n_pairs = 3000\n", + " idx_a = torch.randint(0, 1000, (n_pairs,))\n", + " idx_b = torch.randint(0, 1000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1\n", + " labels[pair_cos <= t1] = 2\n", + " enriched_a, _ = bank(embs[idx_a])\n", + " enriched_b, _ = bank(embs[idx_b])\n", + "\n", + " for mode in [\"with_bank\", \"without_bank\"]:\n", + " if mode == \"with_bank\":\n", + " feat_dim = (768 + 128) * 2\n", + " features = torch.cat([enriched_a, enriched_b], dim=-1)\n", + " else:\n", + " feat_dim = 768 * 2\n", + " features = torch.cat([embs[idx_a], embs[idx_b]], dim=-1)\n", + "\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, 256), nn.GELU(), nn.LayerNorm(256),\n", + " nn.Linear(256, 3)\n", + " ).to(DEVICE)\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " n_clf_train = 2400\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + " for e in range(30):\n", + " clf.train()\n", + " logits = clf(train_f)\n", + " loss = F.cross_entropy(logits, train_l)\n", + " loss.backward(); clf_opt.step(); clf_opt.zero_grad()\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " v_acc = (clf(val_f).argmax(-1) == val_l).float().mean().item()\n", + " t_acc = (clf(train_f).argmax(-1) == train_l).float().mean().item()\n", + " print(f\" {mode:15s}: train={t_acc:.3f} val={v_acc:.3f} gap={t_acc-v_acc:.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Consensus CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Consensus eff_dim:{consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Student v_cos: {v_cos:.3f}\")\n", + " print(f\" Student v_cv: {v_cv:.3f}\")\n", + " print(f\" Bank params: {bank_params:,}\")\n", + " print(f\" Bank geo_eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Bank geo_cv: {geo_cv:.4f}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "c3c92d93759d4f1bacf7b0c9e82fea8c", + "d81ed69dac29481098abe7e3367dd946", + "eae680c132324913a1bd220bc99343b1", + "a3ae20ce4d5d4a8cb5aaa2792f4ac8fb", + "d969fd9386254c979c9195b7ba4fc9da", + "556728e1ff704dcd8b8110d445984172", + "7f0c93c28b2c4b529e15a5baa045c9f2", + "7c5142b27e8042a1861a87fb880a3ffd", + "1f1255df608f4d3eaec9a2d618a73c2e", + "d4cc4e02fcf4457d9c7641457c33851e", + "9ba7c36a314741cf837f9fa47e3c4d4e", + "9ee9cc7e622641d19fc3d3933f01f69d", + "c74ba5f763d94eb8803dfbf92ffe42b7", + "ba31d5f5e2614febb1d312c240f7ff3a", + "c0391a88ed0940b48f567e11db716d61", + "937863ac9ff4471797515158207b6ef6", + "e3dcb07e408d441f926c7778dc0a3bba", + "b11dbe41f79e4e178ef2e5bc75271414", + "43c9d5efd89a4a5ea4ffeb70c55ed556", + "a15366153a1a4dd1a034173435d3099e", + "23600aa3bd934c5abe36d4ea031ae839", + "51e4afb3f3cc4698ada6e1ae1de6bae9" + ] + }, + "id": "wVSD90Ad7GkK", + "outputId": "63dee4d2-7c1c-42a1-8cf1-c258f1df543e" + }, + "execution_count": 6, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "RAPID PROTOTYPE v2: Differentiation-Centered Bank\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "PHASE 0: EXTRACTION\n", + "=================================================================\n", + " Captions: 20,000\n", + "\n", + " Extracting: bert...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00 0:\n", + " aux[\"cross_expert_var\"] = cross_features.var(dim=0).mean()\n", + " else:\n", + " aux[\"cross_expert_var\"] = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # 6. Disagreement preservation\n", + " # The distribution of disagreement should stay at the measured target\n", + " batch_cross_mean = cross_features.mean() if cross_features.shape[1] > 0 else torch.tensor(0.0, device=emb.device)\n", + " batch_cross_std = cross_features.std() if cross_features.shape[1] > 0 else torch.tensor(0.0, device=emb.device)\n", + " batch_disagree_ratio = disagreement_ratio.mean()\n", + " aux[\"disagree_preserve\"] = (\n", + " (batch_cross_mean - self.target_cross_cos_mean).pow(2) +\n", + " (batch_cross_std - self.target_cross_cos_std).pow(2) +\n", + " (batch_disagree_ratio - self.target_disagreement_ratio).pow(2)\n", + " )\n", + "\n", + " # 7. Bank CV\n", + " if B >= 10:\n", + " ctx_n = F.normalize(geo_context, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = ctx_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " bank_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"bank_cv\"] = bank_cv\n", + " else:\n", + " aux[\"bank_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # 8. Emb CV\n", + " if B >= 10:\n", + " emb_n = F.normalize(emb, dim=-1)\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=embedding.device)[:5]\n", + " pts = emb_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " emb_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " aux[\"emb_cv\"] = emb_cv\n", + " else:\n", + " aux[\"emb_cv\"] = torch.tensor(0.0, device=embedding.device)\n", + "\n", + " # Diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"anchor_mean_cos\"] = anchor_cos.mean().item()\n", + " if cross_features.shape[1] > 0:\n", + " aux[\"cross_expert_cos\"] = cross_features.mean().item()\n", + " aux[\"cross_expert_cos_std\"] = cross_features.std().item()\n", + " aux[\"disagreement_ratio\"] = disagreement_ratio.mean().item()\n", + " aux[\"norm_ratio_spread\"] = norm_ratio.std(dim=-1).mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux):\n", + " \"\"\"All targets from measured consensus. Preserves disagreement structure.\"\"\"\n", + " loss = (\n", + " 1.0 * aux[\"expert_agreement\"] +\n", + " 1.0 * aux[\"rotation_ortho\"] +\n", + " 0.5 * aux[\"anchor_spread\"] +\n", + " 0.1 * aux[\"anchor_entropy\"] +\n", + " 0.3 * aux[\"cross_expert_var\"] +\n", + " 0.3 * (aux[\"bank_cv\"] - self.target_cv).abs() +\n", + " 0.3 * (aux[\"emb_cv\"] - self.target_cv).abs() +\n", + " 0.5 * aux[\"disagree_preserve\"] # preserve the disagreement distribution\n", + " )\n", + " return loss\n", + "\n", + " @torch.no_grad()\n", + " def calibrate_disagreement(self, embeddings):\n", + " \"\"\"\n", + " Measure the initial disagreement structure from per-sample distribution.\n", + " Uses the full batch to capture the spread, not just the mean.\n", + " \"\"\"\n", + " B = embeddings.shape[0]\n", + " emb = embeddings.float()\n", + "\n", + " # Compute per-sample disagreement directly\n", + " per_sample_expert_cos = []\n", + " for i in range(self.n_experts):\n", + " R = self.expert_rotations[i]\n", + " W = self.expert_whiteners[i]\n", + " mu = self.expert_means[i]\n", + " centered = emb - mu\n", + " whitened = centered @ W\n", + " whitened_n = F.normalize(whitened, dim=-1)\n", + " in_expert = whitened_n @ R.T\n", + " back = in_expert @ R\n", + " cos = F.cosine_similarity(whitened_n, back, dim=-1)\n", + " per_sample_expert_cos.append(cos)\n", + "\n", + " expert_cos = torch.stack(per_sample_expert_cos, dim=-1) # (B, n_experts)\n", + " per_sample_agreement = expert_cos.mean(dim=-1)\n", + " per_sample_disagreement = expert_cos.std(dim=-1)\n", + " per_sample_ratio = per_sample_disagreement / (per_sample_agreement + 1e-8)\n", + "\n", + " # Cross-expert cosines\n", + " cross_vals = []\n", + " expert_projected = []\n", + " for i in range(self.n_experts):\n", + " R = self.expert_rotations[i]\n", + " W = self.expert_whiteners[i]\n", + " mu = self.expert_means[i]\n", + " centered = emb - mu\n", + " whitened = centered @ W\n", + " whitened_n = F.normalize(whitened, dim=-1)\n", + " expert_projected.append(whitened_n @ R.T)\n", + "\n", + " for i in range(self.n_experts):\n", + " for j in range(i + 1, self.n_experts):\n", + " cc = F.cosine_similarity(expert_projected[i], expert_projected[j], dim=-1)\n", + " cross_vals.append(cc)\n", + "\n", + " if cross_vals:\n", + " cross_all = torch.stack(cross_vals, dim=-1)\n", + " self.target_cross_cos_mean.fill_(cross_all.mean().item())\n", + " self.target_cross_cos_std.fill_(cross_all.std().item())\n", + "\n", + " # Use MEDIAN of per-sample ratio (robust to outliers)\n", + " self.target_disagreement_ratio.fill_(per_sample_ratio.median().item())\n", + "\n", + " print(f\" Calibrated disagreement (n={B}):\")\n", + " print(f\" cross_cos: {self.target_cross_cos_mean.item():.4f} ± {self.target_cross_cos_std.item():.4f}\")\n", + " print(f\" disagree_ratio: median={self.target_disagreement_ratio.item():.6f} \"\n", + " f\"mean={per_sample_ratio.mean().item():.6f} \"\n", + " f\"std={per_sample_ratio.std().item():.6f}\")\n", + " print(f\" expert_cos: {expert_cos.mean().item():.4f} ± {expert_cos.std().item():.4f}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# GEOMETRY\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "def cayley_menger_vol2(pts):\n", + " pts = pts.float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " B, V, _ = d2.shape\n", + " cm = torch.zeros(B, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " return s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + "\n", + "def cv_loss(emb, target=0.12, n_samples=16):\n", + " B = emb.shape[0]\n", + " if B < 5: return torch.tensor(0.0, device=emb.device)\n", + " vols = []\n", + " for _ in range(n_samples):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " return (cv - target).abs()\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " v2 = cayley_menger_vol2(emb[idx].unsqueeze(0))\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def measure_consensus_stats(consensus_embs, n_check=2000):\n", + " \"\"\"Measure exact geometric statistics of the consensus manifold.\"\"\"\n", + " embs = consensus_embs[:n_check].float()\n", + " # CV\n", + " cv = cv_metric(embs.to(DEVICE))\n", + " # Pairwise cosine\n", + " sim = embs @ embs.T\n", + " mask = ~torch.eye(embs.shape[0], dtype=torch.bool)\n", + " pairwise = sim[mask]\n", + " mean_cos = pairwise.mean().item()\n", + " # Spectral\n", + " centered = embs - embs.mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered)\n", + " S_norm = (S / (S.sum() + 1e-8)).tolist()[:50]\n", + " # Eff dim\n", + " eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " return {\n", + " \"cv\": cv,\n", + " \"mean_cos\": mean_cos,\n", + " \"spectral\": S_norm,\n", + " \"eff_dim\": eff_dim,\n", + " }\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# EXTRACTION + ALIGNMENT\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=5000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " N_SAMPLES = 20000\n", + " MAX_LEN = 128\n", + " BATCH = 256\n", + "\n", + " # ── Phase 0: Extract ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXTRACTION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_SAMPLES:\n", + " break\n", + " print(f\" Captions: {len(captions):,}\")\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " print(f\"\\n Extracting: {short}...\")\n", + " model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = tokenizer(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + " embeds[short] = torch.cat(all_emb)\n", + " print(f\" Shape: {embeds[short].shape}\")\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 0b: Align + Consensus + Measure ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0b: GENERALIZED PROCRUSTES ALIGNMENT (no reference bias)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " names = [s for _, s, _ in EXPERTS]\n", + "\n", + " # Generalized Procrustes: iteratively align all to their mean\n", + " # No expert is the reference. The centerpoint emerges.\n", + " GPA_ITERS = 10\n", + " current = {name: embeds[name].float() for name in names}\n", + "\n", + " for gpa_iter in range(GPA_ITERS):\n", + " # Compute mean shape\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + "\n", + " # Align each to mean\n", + " new_current = {}\n", + " total_delta = 0.0\n", + " for name in names:\n", + " info = procrustes_align(current[name], mean_shape)\n", + " new_current[name] = apply_align(current[name], info)\n", + " # Measure how much this iteration changed things\n", + " delta = (new_current[name] - current[name]).pow(2).mean().item()\n", + " total_delta += delta\n", + "\n", + " current = new_current\n", + " if gpa_iter == 0 or (gpa_iter + 1) % 3 == 0 or total_delta < 1e-8:\n", + " print(f\" GPA iter {gpa_iter+1}: delta={total_delta:.8f}\")\n", + " if total_delta < 1e-8:\n", + " print(f\" Converged at iteration {gpa_iter+1}\")\n", + " break\n", + "\n", + " # Final alignment: align each expert to the converged mean\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name], mean_shape)\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name], info)\n", + " cos = F.cosine_similarity(\n", + " aligned[name][:2000], mean_shape[:2000], dim=-1).mean().item()\n", + " print(f\" {name:10s}: cos_after={info['cos_after']:.4f} cos_to_mean={cos:.4f}\")\n", + "\n", + " # Consensus = normalized centroid (now equidistant from all experts)\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " for name in names:\n", + " cos = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {cos:.4f}\")\n", + "\n", + " # Verify equidistance\n", + " expert_cos_to_consensus = []\n", + " for name in names:\n", + " c = F.cosine_similarity(consensus[:2000], aligned[name][:2000], dim=-1).mean().item()\n", + " expert_cos_to_consensus.append(c)\n", + " equidist_range = max(expert_cos_to_consensus) - min(expert_cos_to_consensus)\n", + " print(f\" Equidistance range: {equidist_range:.4f} (should be near 0)\")\n", + "\n", + " # Measure EXACT consensus statistics\n", + " print(f\"\\n Measuring consensus statistics...\")\n", + " consensus_stats = measure_consensus_stats(consensus)\n", + " print(f\" CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Mean cos: {consensus_stats['mean_cos']:.4f}\")\n", + " print(f\" Eff dim: {consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Spectral: [{', '.join(f'{s:.4f}' for s in consensus_stats['spectral'][:5])}...]\")\n", + "\n", + " del embeds, aligned\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 1: Train Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: TRAIN STUDENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " tokenizer = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " tokens = tokenizer(captions, max_length=MAX_LEN, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N_SAMPLES - 2000\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:].to(DEVICE)\n", + " val_mask = attention_mask[n_train:].to(DEVICE)\n", + " val_targets = consensus[n_train:].to(DEVICE)\n", + "\n", + " student = MiniStudent(\n", + " vocab_size=tokenizer.vocab_size, max_len=MAX_LEN,\n", + " d_model=256, n_heads=8, n_layers=8, d_ff=1024,\n", + " output_dim=768, dropout=0.1, pad_token_id=tokenizer.pad_token_id\n", + " ).to(DEVICE)\n", + " n_params = sum(p.numel() for p in student.parameters())\n", + " print(f\" Student: {n_params:,} params\")\n", + " print(f\" CV target: {consensus_stats['cv']:.4f}\")\n", + "\n", + " optimizer = torch.optim.AdamW(student.parameters(), lr=3e-4, weight_decay=0.01)\n", + "\n", + " for epoch in range(10):\n", + " student.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " t_loss, t_acc, t_cos, n = 0, 0, 0, 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + " emb = student(train_ids[idx], train_mask[idx])\n", + " tgt = train_targets[idx]\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_stats[\"cv\"])\n", + " loss = l_nce + l_mse + 0.1 * l_cv\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(student.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " t_loss += loss.item(); t_acc += acc; t_cos += cos; n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + " student.eval()\n", + " with torch.no_grad():\n", + " v_emb = student(val_ids, val_mask)\n", + " _, v_acc = infonce(v_emb[:1000], val_targets[:1000])\n", + " v_cos = F.cosine_similarity(v_emb, val_targets, dim=-1).mean().item()\n", + " v_cv = cv_metric(v_emb[:1000])\n", + " print(f\" E{epoch+1}: {elapsed:.0f}s loss={t_loss/d:.4f} \"\n", + " f\"t_acc={t_acc/d:.3f} t_cos={t_cos/d:.3f} \"\n", + " f\"v_acc={v_acc:.3f} v_cos={v_cos:.3f} v_cv={v_cv:.3f}\")\n", + "\n", + " torch.save(student.state_dict(), \"mini_student.pt\")\n", + " print(f\"\\n Student saved. v_cos={v_cos:.3f}, v_cv={v_cv:.3f}\")\n", + "\n", + " # ── Phase 2: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: TRAIN ALIGNMENT BANK (student frozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " student.eval()\n", + " for p in student.parameters():\n", + " p.requires_grad = False\n", + "\n", + " print(\" Pre-encoding through frozen student...\")\n", + " with torch.no_grad():\n", + " all_embs = []\n", + " for i in range(0, n_train, 512):\n", + " j = min(i + 512, n_train)\n", + " emb = student(train_ids[i:j], train_mask[i:j])\n", + " all_embs.append(emb)\n", + " student_embs = torch.cat(all_embs)\n", + " val_student_embs = student(val_ids, val_mask)\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=512, d_bank=128\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names,\n", + " consensus[:n_train], consensus_stats)\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + " print(f\" Bank targets: CV={bank.target_cv.item():.4f}, \"\n", + " f\"mean_cos={bank.target_mean_cos.item():.4f}\")\n", + "\n", + " # Calibrate disagreement from initial state (before any training)\n", + " bank.calibrate_disagreement(student_embs[:2000])\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=1e-3, weight_decay=0.01)\n", + " BANK_EPOCHS = 20\n", + " BANK_BATCH = 256\n", + "\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0,\n", + " \"anchor_spread\": 0, \"bank_cv\": 0, \"emb_cv\": 0,\n", + " \"cross_expert_var\": 0, \"disagree_preserve\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + " emb = student_embs[idx]\n", + " enriched, aux = bank(emb)\n", + " loss = bank.bank_loss(aux)\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " v_enriched, v_aux = bank(val_student_embs)\n", + " v_loss = bank.bank_loss(v_aux).item()\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f}\")\n", + " print(f\" Geometry: b_cv={stats['bank_cv']/d:.4f} e_cv={stats['emb_cv']/d:.4f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} a_max={v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Experts: cos={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f} \"\n", + " f\"agr={stats['expert_agreement']/d:.6f} ortho={stats['rotation_ortho']/d:.6f}\")\n", + " print(f\" Disagree: x_cos={v_aux.get('cross_expert_cos', 0):.4f}±{v_aux.get('cross_expert_cos_std', 0):.4f} \"\n", + " f\"ratio={v_aux['disagreement_ratio']:.6f} \"\n", + " f\"preserve={stats['disagree_preserve']/d:.6f} \"\n", + " f\"norms={v_aux['norm_ratio_spread']:.4f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank.pt\")\n", + "\n", + " # ── Phase 3: Geometric Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " enriched_val, v_aux = bank(val_student_embs)\n", + " original_768 = enriched_val[:, :768]\n", + " geo_context = enriched_val[:, 768:]\n", + "\n", + " passthrough_cos = F.cosine_similarity(\n", + " original_768[:100], val_student_embs[:100], dim=-1).mean().item()\n", + " geo_cv = cv_metric(F.normalize(geo_context[:1000], dim=-1))\n", + " S = torch.linalg.svdvals(\n", + " geo_context[:1000].float() - geo_context[:1000].float().mean(0))\n", + " geo_eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + "\n", + " # Verify consensus stats are preserved\n", + " emb_cv = cv_metric(val_student_embs[:1000])\n", + "\n", + " print(f\" Passthrough: {passthrough_cos:.6f} (target: 1.000)\")\n", + " print(f\" Emb CV: {emb_cv:.4f} (consensus: {consensus_stats['cv']:.4f})\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo eff_dim: {geo_eff_dim:.1f} / {bank.d_bank}\")\n", + " print(f\" Expert cos: {v_aux['expert_cos_mean']:.3f} ± {v_aux['expert_cos_std']:.3f}\")\n", + " print(f\" Anchor max cos: {v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Disagreement:\")\n", + " print(f\" Cross-expert: {v_aux.get('cross_expert_cos', 0):.4f} ± {v_aux.get('cross_expert_cos_std', 0):.4f}\")\n", + " print(f\" Ratio: {v_aux['disagreement_ratio']:.6f} (target: {bank.target_disagreement_ratio.item():.6f})\")\n", + " print(f\" Norm spread: {v_aux['norm_ratio_spread']:.4f}\")\n", + "\n", + " # ── Phase 4: Classifier Stability Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " with torch.no_grad():\n", + " embs = val_student_embs[:1000]\n", + " sim = embs @ embs.T\n", + " sim.fill_diagonal_(-1)\n", + " n_pairs = 3000\n", + " idx_a = torch.randint(0, 1000, (n_pairs,))\n", + " idx_b = torch.randint(0, 1000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1\n", + " labels[pair_cos <= t1] = 2\n", + "\n", + " enriched_a, aux_a = bank(embs[idx_a])\n", + " enriched_b, aux_b = bank(embs[idx_b])\n", + "\n", + " # Build explicit geometric features per pair\n", + " # These are interpretable and hard to overfit\n", + " a_emb = embs[idx_a]; b_emb = embs[idx_b]\n", + " a_geo = enriched_a[:, 768:]; b_geo = enriched_b[:, 768:]\n", + "\n", + " geo_explicit = torch.cat([\n", + " # Pair-level\n", + " F.cosine_similarity(a_emb, b_emb, dim=-1).unsqueeze(-1), # raw cosine\n", + " (a_emb - b_emb).pow(2).mean(dim=-1).unsqueeze(-1), # MSE\n", + " F.cosine_similarity(a_geo, b_geo, dim=-1).unsqueeze(-1), # geo context cosine\n", + " (a_geo - b_geo).pow(2).mean(dim=-1).unsqueeze(-1), # geo context MSE\n", + " # Per-sample bank diagnostics (already computed in forward)\n", + " torch.abs(a_emb - b_emb).mean(dim=-1).unsqueeze(-1), # L1 distance\n", + " (a_emb * b_emb).sum(dim=-1).unsqueeze(-1), # dot product\n", + " ], dim=-1) # (n_pairs, 6)\n", + "\n", + " modes = {\n", + " \"raw_768\": torch.cat([a_emb, b_emb], dim=-1),\n", + " \"raw+diff\": torch.cat([a_emb, b_emb, torch.abs(a_emb - b_emb), a_emb * b_emb], dim=-1),\n", + " \"bank_enriched\": torch.cat([enriched_a, enriched_b], dim=-1),\n", + " \"bank+diff\": torch.cat([enriched_a, enriched_b,\n", + " torch.abs(enriched_a - enriched_b),\n", + " enriched_a * enriched_b], dim=-1),\n", + " \"geo_explicit\": geo_explicit,\n", + " }\n", + "\n", + " print(f\"\\n {'Mode':<20} {'Dim':>6} {'Train':>7} {'Val':>7} {'Gap':>7}\")\n", + " print(f\" {'-'*50}\")\n", + "\n", + " for mode_name, features in modes.items():\n", + " feat_dim = features.shape[1]\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, min(256, feat_dim)), nn.GELU(), nn.LayerNorm(min(256, feat_dim)),\n", + " nn.Dropout(0.1),\n", + " nn.Linear(min(256, feat_dim), 3)\n", + " ).to(DEVICE)\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " n_clf_train = 2400\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + " for e in range(30):\n", + " clf.train()\n", + " logits = clf(train_f)\n", + " loss = F.cross_entropy(logits, train_l)\n", + " loss.backward(); clf_opt.step(); clf_opt.zero_grad()\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " v_acc = (clf(val_f).argmax(-1) == val_l).float().mean().item()\n", + " t_acc = (clf(train_f).argmax(-1) == train_l).float().mean().item()\n", + " print(f\" {mode_name:<20} {feat_dim:>6} {t_acc:>7.3f} {v_acc:>7.3f} {t_acc-v_acc:>7.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Consensus CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Consensus eff_dim:{consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Student v_cos: {v_cos:.3f}\")\n", + " print(f\" Student v_cv: {v_cv:.3f}\")\n", + " print(f\" Bank params: {bank_params:,}\")\n", + " print(f\" Bank geo_eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Bank geo_cv: {geo_cv:.4f}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "adf5073645bd4927b23e9e2f245e57bc", + "41fa9a998197421db2e37852bd643d20", + "3c28a35ed90345d5983b6e7910e0076e", + "19460177f35546d7af56905bfa79dbe9", + "609d8ace2dca4f9aa239ece209512e3c", + "4d1db66017314a47917b4f7e370b2da3", + "78512c07da324a4d91baff786f68f79b", + "e6e0a63481964fbe982e006f2d855bbb", + "ebcea318e84145b6a5d238df793755da", + "a706ddc178fa405a9a058283431ae4a7", + "536780f5538e4c148d6c08014a76f670", + "8acc67491474462e9e062baca52c04a2", + "2eb6770aee654f388176238f09778b30", + "95b5c43c48594913ae0d8e315a9c80f9", + "fb6a8aa0a91b4bb0adc448ad84b1428b", + "2e96ea3acbad4fee9d528107a1e527af", + "aa01aa0a79e1496c82c8a0278cf9c3da", + "3440899e994a4b68971ad5924d9db1ad", + "e761b132c88240549199f12afaf5a650", + "cb3b9be5979a4f9cb8a8186d1e324cdc", + "b343f92042b94bc78eb0af35fd65a1bf", + "4d73be6b0016448da67fa630570807f2" + ] + }, + "id": "MucNEUth94EL", + "outputId": "4e982950-dd69-4ff6-8338-4ee3540c05c4" + }, + "execution_count": 9, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "RAPID PROTOTYPE v2: Differentiation-Centered Bank\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "PHASE 0: EXTRACTION\n", + "=================================================================\n", + " Captions: 20,000\n", + "\n", + " Extracting: bert...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00= 10:\n", + " vols = []\n", + " for _ in range(32):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " pts = data[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " aux[label] = stacked.std() / (stacked.mean() + 1e-8)\n", + " else:\n", + " aux[label] = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # Diagnostics\n", + " aux[\"expert_cos_mean\"] = expert_cos.mean().item()\n", + " aux[\"expert_cos_std\"] = expert_cos.std().item()\n", + " aux[\"anchor_max_cos\"] = anchor_cos.max(dim=-1).values.mean().item()\n", + " aux[\"cross_expert_cos\"] = cross_features.mean().item()\n", + " aux[\"cross_expert_cos_std\"] = cross_features.std().item()\n", + " aux[\"disagreement_ratio\"] = disagreement_ratio.mean().item()\n", + " aux[\"norm_ratio_spread\"] = norm_ratio.std(dim=-1).mean().item()\n", + "\n", + " return enriched, aux\n", + "\n", + " def bank_loss(self, aux):\n", + " return (\n", + " 1.0 * aux[\"expert_agreement\"] +\n", + " 1.0 * aux[\"rotation_ortho\"] +\n", + " 0.5 * aux[\"anchor_spread\"] +\n", + " 0.1 * aux[\"anchor_entropy\"] +\n", + " 0.3 * aux[\"cross_expert_var\"] +\n", + " 0.3 * (aux[\"bank_cv\"] - self.target_cv).abs() +\n", + " 0.3 * (aux[\"emb_cv\"] - self.target_cv).abs() +\n", + " 0.5 * aux[\"disagree_preserve\"])\n", + "\n", + " @torch.no_grad()\n", + " def calibrate_disagreement(self, embeddings):\n", + " B = embeddings.shape[0]\n", + " emb = embeddings.float()\n", + " per_sample_expert_cos = []\n", + " expert_projected = []\n", + " for i in range(self.n_experts):\n", + " R = self.expert_rotations[i]; W = self.expert_whiteners[i]; mu = self.expert_means[i]\n", + " centered = emb - mu; whitened = centered @ W\n", + " whitened_n = F.normalize(whitened, dim=-1)\n", + " in_expert = whitened_n @ R.T\n", + " back = in_expert @ R\n", + " per_sample_expert_cos.append(F.cosine_similarity(whitened_n, back, dim=-1))\n", + " expert_projected.append(in_expert)\n", + " expert_cos = torch.stack(per_sample_expert_cos, dim=-1)\n", + " per_sample_ratio = expert_cos.std(dim=-1) / (expert_cos.mean(dim=-1) + 1e-8)\n", + " cross_vals = []\n", + " for i in range(self.n_experts):\n", + " for j in range(i + 1, self.n_experts):\n", + " cross_vals.append(F.cosine_similarity(expert_projected[i], expert_projected[j], dim=-1))\n", + " cross_all = torch.stack(cross_vals, dim=-1)\n", + " self.target_cross_cos_mean.fill_(cross_all.mean().item())\n", + " self.target_cross_cos_std.fill_(cross_all.std().item())\n", + " self.target_disagreement_ratio.fill_(per_sample_ratio.median().item())\n", + " print(f\" Calibrated (n={B}):\")\n", + " print(f\" cross_cos: {self.target_cross_cos_mean.item():.4f} ± {self.target_cross_cos_std.item():.4f}\")\n", + " print(f\" disagree_ratio: median={self.target_disagreement_ratio.item():.6f}\")\n", + " print(f\" expert_cos: {expert_cos.mean().item():.4f} ± {expert_cos.std().item():.4f}\")\n", + " print(f\" cross pairs: {len(cross_vals)}\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# ALIGNMENT UTILITIES\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=10000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\n", + " \"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after,\n", + " }\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "def cv_metric(emb, n=200):\n", + " B = emb.shape[0]\n", + " if B < 5: return 0.0\n", + " vols = []\n", + " for _ in range(n):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " pts = emb[idx].unsqueeze(0).float()\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " v = torch.sqrt(F.relu(v2[0]) + 1e-12).item()\n", + " if v > 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " names = [s for _, s, _ in EXPERTS]\n", + "\n", + " # ── Phase 0: Extract or Load Embeddings ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: EXPERT EMBEDDINGS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " os.makedirs(CACHE_DIR, exist_ok=True)\n", + " caps_path = os.path.join(CACHE_DIR, \"captions.json\")\n", + "\n", + " # Check what's cached\n", + " all_cached = all(\n", + " os.path.exists(os.path.join(CACHE_DIR, f\"{s}.pt\"))\n", + " for _, s, _ in EXPERTS)\n", + "\n", + " if all_cached:\n", + " print(\" Loading cached embeddings...\")\n", + " embeds = {}\n", + " for _, short, _ in EXPERTS:\n", + " embeds[short] = torch.load(\n", + " os.path.join(CACHE_DIR, f\"{short}.pt\"), weights_only=True)\n", + " print(f\" {short}: {embeds[short].shape}\")\n", + " if os.path.exists(caps_path):\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " print(f\" Captions: {len(captions):,}\")\n", + " else:\n", + " print(\" captions.json missing, reloading...\")\n", + " from datasets import load_dataset\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_EXTRACT:\n", + " break\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + " else:\n", + " # Extract from scratch\n", + " from datasets import load_dataset\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " print(f\" Loading {N_EXTRACT:,} captions...\")\n", + " ds = load_dataset(\"CaptionEmporium/conceptual-captions-cc12m-llavanext\",\n", + " split=\"train\", streaming=True)\n", + " captions = []\n", + " for row in ds:\n", + " cap = row.get(\"caption_llava\", \"\")\n", + " if isinstance(cap, str) and len(cap) > 50:\n", + " captions.append(cap)\n", + " if len(captions) >= N_EXTRACT:\n", + " break\n", + " print(f\" Got {len(captions):,} captions\")\n", + "\n", + " with open(caps_path, \"w\") as f:\n", + " json.dump(captions, f)\n", + "\n", + " embeds = {}\n", + " for model_name, short, max_len in EXPERTS:\n", + " out_path = os.path.join(CACHE_DIR, f\"{short}.pt\")\n", + " if os.path.exists(out_path):\n", + " embeds[short] = torch.load(out_path, weights_only=True)\n", + " print(f\" {short}: cached {embeds[short].shape}\")\n", + " continue\n", + "\n", + " print(f\"\\n Extracting: {short} ({model_name}, max_len={max_len})...\")\n", + " ext_model = AutoModel.from_pretrained(model_name).to(DEVICE).eval()\n", + " ext_tok = AutoTokenizer.from_pretrained(model_name)\n", + " n_p = sum(p.numel() for p in ext_model.parameters())\n", + " print(f\" {n_p:,} params\")\n", + "\n", + " all_emb = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, len(captions), 128), desc=f\" {short}\"):\n", + " batch = captions[i:i+128]\n", + " inputs = ext_tok(batch, max_length=max_len, padding=True,\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = ext_model(**inputs)\n", + " m = inputs.attention_mask.unsqueeze(-1).float()\n", + " pooled = (out.last_hidden_state * m).sum(1) / m.sum(1).clamp(min=1)\n", + " all_emb.append(pooled.cpu())\n", + "\n", + " emb = torch.cat(all_emb)\n", + " if emb.shape[1] != 768:\n", + " emb = emb[:, :768] if emb.shape[1] > 768 else F.pad(emb, (0, 768 - emb.shape[1]))\n", + " embeds[short] = emb\n", + " torch.save(emb, out_path)\n", + " print(f\" Saved: {emb.shape}\")\n", + " del ext_model, ext_tok; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " N = min(len(captions), min(e.shape[0] for e in embeds.values()))\n", + " print(f\" Using {N:,} samples\")\n", + "\n", + " # ── Phase 1: GPA Alignment ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: GENERALIZED PROCRUSTES ALIGNMENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " current = {name: embeds[name][:N].float() for name in names}\n", + " for gpa_iter in range(15):\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + " total_delta = 0.0\n", + " new_current = {}\n", + " for name in names:\n", + " info = procrustes_align(current[name], mean_shape)\n", + " new_current[name] = apply_align(current[name], info)\n", + " total_delta += (new_current[name] - current[name]).pow(2).mean().item()\n", + " current = new_current\n", + " if gpa_iter == 0 or (gpa_iter + 1) % 3 == 0:\n", + " print(f\" GPA iter {gpa_iter+1}: delta={total_delta:.8f}\")\n", + " if total_delta < 1e-8:\n", + " print(f\" Converged at iteration {gpa_iter+1}\")\n", + " break\n", + "\n", + " # Final alignment to converged mean\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + " procrustes_results = {}\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name][:N], mean_shape)\n", + " procrustes_results[name] = info\n", + " aligned[name] = apply_align(embeds[name][:N], info)\n", + " cos_to_mean = F.cosine_similarity(\n", + " aligned[name][:5000], mean_shape[:5000], dim=-1).mean().item()\n", + " print(f\" {name:10s}: cos_after={info['cos_after']:.4f} cos_to_mean={cos_to_mean:.4f}\")\n", + "\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " expert_cos_to_consensus = []\n", + " for name in names:\n", + " c = F.cosine_similarity(consensus[:5000], aligned[name][:5000], dim=-1).mean().item()\n", + " expert_cos_to_consensus.append(c)\n", + " print(f\" cos(consensus, {name}): {c:.4f}\")\n", + " equidist = max(expert_cos_to_consensus) - min(expert_cos_to_consensus)\n", + " print(f\" Equidistance range: {equidist:.4f}\")\n", + "\n", + " # Measure consensus statistics\n", + " print(f\"\\n Measuring consensus statistics...\")\n", + " c_sub = consensus[:5000].to(DEVICE)\n", + " consensus_cv = cv_metric(c_sub)\n", + " sim = c_sub @ c_sub.T\n", + " mask = ~torch.eye(5000, dtype=torch.bool, device=DEVICE)\n", + " mean_cos = sim[mask].mean().item()\n", + " centered = c_sub.float() - c_sub.float().mean(0, keepdim=True)\n", + " S = torch.linalg.svdvals(centered)\n", + " spectral = (S / (S.sum() + 1e-8)).cpu().tolist()[:50]\n", + " eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + " consensus_stats = {\"cv\": consensus_cv, \"mean_cos\": mean_cos,\n", + " \"spectral\": spectral, \"eff_dim\": eff_dim}\n", + " print(f\" CV: {consensus_cv:.4f}\")\n", + " print(f\" Mean cos: {mean_cos:.4f}\")\n", + " print(f\" Eff dim: {eff_dim:.1f}\")\n", + " del c_sub, sim; torch.cuda.empty_cache()\n", + "\n", + " del embeds, aligned, current, mean_shape\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 2: Load + Encode Frozen Student ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: ENCODE FROZEN STUDENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + " for p in model.parameters():\n", + " p.requires_grad = False\n", + " print(f\" Student: {sum(p.numel() for p in model.parameters()):,} params (frozen)\")\n", + "\n", + " # captions already loaded from Phase 0\n", + " captions = captions[:N]\n", + " print(f\" Encoding {N:,} captions...\")\n", + " all_student_embs = []\n", + " with torch.no_grad():\n", + " for i in tqdm(range(0, N, 256), desc=\" Encoding\"):\n", + " j = min(i + 256, N)\n", + " inputs = tokenizer(captions[i:j], max_length=512, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " out = model(**inputs)\n", + " all_student_embs.append(out.last_hidden_state.cpu())\n", + " student_embs = torch.cat(all_student_embs).to(DEVICE)\n", + " print(f\" Student embeddings: {student_embs.shape}\")\n", + "\n", + " del model\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # Split\n", + " n_train = N - N_VAL\n", + " train_embs = student_embs[:n_train]\n", + " val_embs = student_embs[n_train:n_train + N_VAL]\n", + " print(f\" Train: {n_train:,} Val: {N_VAL:,}\")\n", + "\n", + " # ── Phase 3: Train Alignment Bank ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: TRAIN ALIGNMENT BANK\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank = AlignmentBank(\n", + " d_embed=768, n_experts=len(EXPERTS),\n", + " n_anchors=N_ANCHORS, d_bank=D_BANK\n", + " ).to(DEVICE)\n", + "\n", + " bank.init_from_procrustes(procrustes_results, names,\n", + " consensus[:n_train], consensus_stats)\n", + " bank.calibrate_disagreement(train_embs[:5000])\n", + "\n", + " bank_params = sum(p.numel() for p in bank.parameters())\n", + " print(f\" Bank: {bank_params:,} params\")\n", + "\n", + " bank_opt = torch.optim.AdamW(bank.parameters(), lr=BANK_LR, weight_decay=0.01)\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " bank_opt, T_max=(n_train // BANK_BATCH) * BANK_EPOCHS, eta_min=1e-5)\n", + "\n", + " best_v_loss = float(\"inf\")\n", + " for epoch in range(BANK_EPOCHS):\n", + " bank.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss = 0\n", + " stats = {\"expert_agreement\": 0, \"rotation_ortho\": 0, \"anchor_spread\": 0,\n", + " \"bank_cv\": 0, \"emb_cv\": 0, \"cross_expert_var\": 0, \"disagree_preserve\": 0}\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BANK_BATCH):\n", + " idx = perm[i:i+BANK_BATCH]\n", + " if len(idx) < 16: continue\n", + " _, aux = bank(train_embs[idx])\n", + " loss = bank.bank_loss(aux)\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(bank.parameters(), 1.0)\n", + " bank_opt.step(); bank_opt.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + " total_loss += loss.item()\n", + " for k in stats:\n", + " if k in aux:\n", + " v = aux[k]\n", + " stats[k] += v.item() if torch.is_tensor(v) else v\n", + " n += 1\n", + "\n", + " elapsed = time.time() - t0; d = max(n, 1)\n", + "\n", + " bank.eval()\n", + " with torch.no_grad():\n", + " _, v_aux = bank(val_embs)\n", + " v_loss = bank.bank_loss(v_aux).item()\n", + "\n", + " if v_loss < best_v_loss:\n", + " best_v_loss = v_loss\n", + " torch.save(bank.state_dict(), \"alignment_bank_best.pt\")\n", + "\n", + " if (epoch + 1) % 5 == 0 or epoch == 0:\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f}\")\n", + " print(f\" Geometry: b_cv={stats['bank_cv']/d:.4f} e_cv={stats['emb_cv']/d:.4f} \"\n", + " f\"spread={stats['anchor_spread']/d:.5f} a_max={v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Experts: cos={v_aux['expert_cos_mean']:.3f}±{v_aux['expert_cos_std']:.3f} \"\n", + " f\"agr={stats['expert_agreement']/d:.6f} ortho={stats['rotation_ortho']/d:.6f}\")\n", + " print(f\" Disagree: x_cos={v_aux['cross_expert_cos']:.4f}±{v_aux['cross_expert_cos_std']:.4f} \"\n", + " f\"ratio={v_aux['disagreement_ratio']:.6f} \"\n", + " f\"preserve={stats['disagree_preserve']/d:.6f}\")\n", + " else:\n", + " print(f\" E{epoch+1:2d}: {elapsed:.0f}s loss={total_loss/d:.4f} v_loss={v_loss:.4f} \"\n", + " f\"exp={v_aux['expert_cos_mean']:.3f} \"\n", + " f\"b_cv={stats['bank_cv']/d:.4f} \"\n", + " f\"x_cos={v_aux['cross_expert_cos']:.4f}\")\n", + "\n", + " torch.save(bank.state_dict(), \"alignment_bank_final.pt\")\n", + "\n", + " # ── Phase 4: Geometric Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: GEOMETRIC VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " bank.load_state_dict(torch.load(\"alignment_bank_best.pt\", weights_only=True))\n", + " bank.eval()\n", + "\n", + " with torch.no_grad():\n", + " enriched_val, v_aux = bank(val_embs)\n", + " original_768 = enriched_val[:, :768]\n", + " geo_context = enriched_val[:, 768:]\n", + "\n", + " passthrough = F.cosine_similarity(\n", + " original_768[:100], val_embs[:100], dim=-1).mean().item()\n", + " geo_cv = cv_metric(F.normalize(geo_context[:2000], dim=-1))\n", + " S = torch.linalg.svdvals(\n", + " geo_context[:2000].float() - geo_context[:2000].float().mean(0))\n", + " geo_eff_dim = float((S.sum() ** 2) / (S.pow(2).sum() + 1e-12))\n", + " emb_cv = cv_metric(val_embs[:2000])\n", + "\n", + " print(f\" Passthrough: {passthrough:.6f}\")\n", + " print(f\" Emb CV: {emb_cv:.4f} (consensus: {consensus_stats['cv']:.4f})\")\n", + " print(f\" Geo context CV: {geo_cv:.4f}\")\n", + " print(f\" Geo eff_dim: {geo_eff_dim:.1f} / {D_BANK}\")\n", + " print(f\" Expert cos: {v_aux['expert_cos_mean']:.3f} ± {v_aux['expert_cos_std']:.3f}\")\n", + " print(f\" Anchor max cos: {v_aux['anchor_max_cos']:.3f}\")\n", + " print(f\" Cross-expert: {v_aux['cross_expert_cos']:.4f} ± {v_aux['cross_expert_cos_std']:.4f}\")\n", + " print(f\" Disagree ratio: {v_aux['disagreement_ratio']:.6f}\")\n", + "\n", + " # ── Phase 5: Classifier Test ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 5: CLASSIFIER STABILITY TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " with torch.no_grad():\n", + " embs = val_embs[:2000]\n", + " sim = embs @ embs.T; sim.fill_diagonal_(-1)\n", + " n_pairs = 5000\n", + " idx_a = torch.randint(0, 2000, (n_pairs,))\n", + " idx_b = torch.randint(0, 2000, (n_pairs,))\n", + " pair_cos = sim[idx_a, idx_b]\n", + " sorted_cos, _ = pair_cos.sort()\n", + " t1 = sorted_cos[n_pairs // 3].item()\n", + " t2 = sorted_cos[2 * n_pairs // 3].item()\n", + " labels = torch.zeros(n_pairs, dtype=torch.long, device=DEVICE)\n", + " labels[pair_cos > t2] = 0\n", + " labels[(pair_cos <= t2) & (pair_cos > t1)] = 1\n", + " labels[pair_cos <= t1] = 2\n", + " enriched_a, _ = bank(embs[idx_a])\n", + " enriched_b, _ = bank(embs[idx_b])\n", + " a_emb = embs[idx_a]; b_emb = embs[idx_b]\n", + " a_geo = enriched_a[:, 768:]; b_geo = enriched_b[:, 768:]\n", + " geo_explicit = torch.cat([\n", + " F.cosine_similarity(a_emb, b_emb, dim=-1).unsqueeze(-1),\n", + " (a_emb - b_emb).pow(2).mean(dim=-1).unsqueeze(-1),\n", + " F.cosine_similarity(a_geo, b_geo, dim=-1).unsqueeze(-1),\n", + " (a_geo - b_geo).pow(2).mean(dim=-1).unsqueeze(-1),\n", + " torch.abs(a_emb - b_emb).mean(dim=-1).unsqueeze(-1),\n", + " (a_emb * b_emb).sum(dim=-1).unsqueeze(-1),\n", + " ], dim=-1)\n", + "\n", + " modes = {\n", + " \"raw_768\": torch.cat([a_emb, b_emb], dim=-1),\n", + " \"raw+diff\": torch.cat([a_emb, b_emb, torch.abs(a_emb - b_emb), a_emb * b_emb], dim=-1),\n", + " \"bank_enriched\": torch.cat([enriched_a, enriched_b], dim=-1),\n", + " \"bank+diff\": torch.cat([enriched_a, enriched_b,\n", + " torch.abs(enriched_a - enriched_b),\n", + " enriched_a * enriched_b], dim=-1),\n", + " \"geo_explicit\": geo_explicit,\n", + " }\n", + "\n", + " print(f\"\\n {'Mode':<20} {'Dim':>6} {'Train':>7} {'Val':>7} {'Gap':>7}\")\n", + " print(f\" {'-'*50}\")\n", + "\n", + " n_clf_train = 4000\n", + " for mode_name, features in modes.items():\n", + " feat_dim = features.shape[1]\n", + " clf = nn.Sequential(\n", + " nn.Linear(feat_dim, min(256, feat_dim)), nn.GELU(),\n", + " nn.LayerNorm(min(256, feat_dim)), nn.Dropout(0.1),\n", + " nn.Linear(min(256, feat_dim), 3)).to(DEVICE)\n", + " clf_opt = torch.optim.Adam(clf.parameters(), lr=1e-3)\n", + " train_f = features[:n_clf_train].detach()\n", + " train_l = labels[:n_clf_train]\n", + " val_f = features[n_clf_train:].detach()\n", + " val_l = labels[n_clf_train:]\n", + " for e in range(30):\n", + " clf.train()\n", + " loss = F.cross_entropy(clf(train_f), train_l)\n", + " loss.backward(); clf_opt.step(); clf_opt.zero_grad()\n", + " clf.eval()\n", + " with torch.no_grad():\n", + " v_acc = (clf(val_f).argmax(-1) == val_l).float().mean().item()\n", + " t_acc = (clf(train_f).argmax(-1) == train_l).float().mean().item()\n", + " print(f\" {mode_name:<20} {feat_dim:>6} {t_acc:>7.3f} {v_acc:>7.3f} {t_acc-v_acc:>7.3f}\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Consensus CV: {consensus_stats['cv']:.4f}\")\n", + " print(f\" Consensus eff_dim: {consensus_stats['eff_dim']:.1f}\")\n", + " print(f\" Equidistance: {equidist:.4f}\")\n", + " print(f\" Bank params: {bank_params:,}\")\n", + " print(f\" Bank geo eff_dim: {geo_eff_dim:.1f}\")\n", + " print(f\" Bank geo CV: {geo_cv:.4f}\")\n", + " print(f\" Best val loss: {best_v_loss:.4f}\")\n", + " print(f\"\\n Files: alignment_bank_best.pt, alignment_bank_final.pt\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "b63e64cfb73141dca1082ac02358b288", + "fa1545986ad94525a89ee64cd3ed2341", + "62da4d748f644d36a45c1a9306c226f2", + "1da92130d7114b55a8a4fc691337691e", + "3e8ee8649aef4331af229d584a6bccba", + "1b06a9738e444782b7885db78dfd30c9", + "1114ece9b9884973a790259962983edb", + "ae2c80f5d3db4b5f8b9748a34ea27906", + "e3c8e44dcd0e4b3b84a5c60493e8684e", + "0fa8053b5f6340a49ed293ca538c41e0", + "00dfa9645b234496be78e8f8e58c8e38", + "2dcc8c0441bb4966a6c9ec3b810c06fe", + "e83df993ae8c4e26a9aef7c2ecafc810", + "370d7642e679416690f7380ae3ba03b4", + "2ece17832e524e80a1740571ea0557f8", + "1ab761ee76294cb1a049a3c672bf47af", + "da9b05cbed414810bab33c39f192f7e8", + "ed090236ac4a42598beb85e2916abe1e", + "85c3b5826c104280bb2a3483f87bcae3", + "5c0e201b8dfb4db4a587d1dc0046089f", + "ca439a1c8d9a42f1b5a3cce4e936b60d", + "b7fea8a491a84fb99d7e1aea9fc7f61f", + "c73d64f12b3547fc8cd59b4598812465", + "602a23aaf0a4401aa5e3c9263f6026c5", + "170719a38be14c138a17a4a9c315104d", + "3e7a63e286344af1ba9a49e7ca55380c", + "a5bb395b6b5c4c6b9008ae7efbe5f879", + "b00700760d5744a3abcecda5c56a6a37", + "7e2d5b90cb674b0aade480ca52539ab9", + "06d1a27817564802bd85b5144a8f66b1", + "c4885af32a6f45e682a15bd7b2681404", + "88a1cc0f868041bf97319dc4eb14965c", + "f78341ea0b984611ab0320ada1474f00", + "3d49c72bf02f46d99659693617d02bad", + "a41e39e89be040b79b23ffa3c90fb41f", + "c2d339a959a94b3b893172e6b79a46f4", + "9a5b2d6e52b34c5e827f271d159ae4d1", + "5a47fce188ba4a54835f9c203b3161e5", + "9d97322824c4451baba172989e0625d1", + "d6fdf04a597045fb80130b54ac706d95", + "78ee2e9a06c84bacb21aba15ffc4c8f4", + "27269f63df6a4e9aa0093279f7b42d08", + "35ad86ea0710430fad41d1dc09c7855e", + "48db0905592b4b2abc3ab5360b00f058", + "7f76e88c6b6b4df294c9e06abff02300", + "fc561d3e7239434f82e5f97834882373", + "39c70d8edbca4f71b171f8d278d3e7da", + "78a65d554d0d4c3b8712c96391798a85", + "9f71e2c496e94633ab5cc7627713d0b9", + "3c6ead09d8384533ad6d6dbc1e092bf7", + "7bb75c4743744fa4980ab172337295c6", + "06f346753e6341708bff1df229606ee7", + "b183d808dc054620bef47aa1abaa3b81", + "34a6088054c24d71b0a5301457f8feb6", + "0979317c03b248eda6f6a22b23201fdb", + "5848fb940af846cd933f3e422531bf8f", + "838f04dc0d9542309882c1b3f4ade428", + "304f366145bc45e2bb18a8cc8e580bd0", + "d33504dcb688479d8bbbafc42a5c175b", + "15302e84f6b14b86be92850a8b64b2b2", + "dae34e28a39d49da94c1f1423fc6f03e", + "95cb9b137b404a81a8a564d269aebfca", + "bbe461e321ed4f15bfb4a415ddaa6b5e", + "33e1e90c5b7d40539cb0d3eff1517215", + "46a74a567a034bbfa01c1a38bf44a560", + "d3e26a070390464bbd887323979f6b9d", + "59edb1de333f438b85b3a1c170ccbb99", + "0254b5bac7ad4839893e165d49b99caa", + "bd869600206c472f9c088e24cca59a06", + "77c11e3cbdd344ea84f99dc532c74ac7", + "ca0c288dfb10404e8d3b006a1cd37547", + "7756944b4f9b472f8dc77e65a0cb28b8", + "2a6eb8cfa49e4be6b63010060f0545d4", + "4cb348841ab0465ab6020d0060446a8a", + "279f3fbfb3174fd7b6072776b3a18c17", + "ce3f9808f34f484dabd51d98aa83c45c", + "6a13a1052fe7464493deab5f18087512", + "97ae1151703b49c2b4552925c6bf146d", + "dee89038279f4bf597b6fee92851846f", + "8a859087e043457fb89e91c1cb7a7518", + "36c55cdd3c594853b839a4fd804b6719", + "aa4b81df2e7c48e9a1fc58768ccab315", + "531ef4668b0244379e244746a5663ac4", + "84101638e6d944dabf39d0fd107ebe8a", + "7f7ac2e9e12648a1b13c92c11e159bd2", + "6bfcbcb3ee5d4feb90366a3f392ed7de", + "1b7cee9b57c04a63b5dde8582963846f", + "b30bcdd589434fba8082f5c145ffba82", + "5025ac720aaa4b0a99300be21c58a094", + "47adec0f7ea64f9284262962d0ad5d67", + "0496e5b26cfa4efead30f5f848bde313", + "5f48f891cd7548d49115b6f87abe9ce2", + "ab79ef07a97c467494f92ad657acd890", + "7f2c27e564f6470285ffc1ad74a858ab", + "342f16b9972749178ffc62d3c5d6f0e5", + "eae405f7ae774cfa952e8550f136ee50", + "1433f037c55246a2b5c62bfea9a7a287", + "6559e7091b1446239f8b320e1f5b3103", + "f4949533579247548a9ef75f12b4126a", + "15cc16bacc074bc59293f5d824dcc8b4", + "343fea9e802d40eb872f1234e4feeb9d", + "bd17bb197fbd4081b1696427b9f4d37a", + "82083d75226140ef9ffa0e568e355a7a", + "603a9a7cdd3c46518f110d0098513e92", + "bdab02de69de42239423c469597386df", + "e0e7903477c944ecb512ffe5f3ff54f1", + "7d453ba92fcb4873aa7f24fa56f0f4ff", + "506ab6234b8a481ebf30a2457c47aba4", + "09b3f082f0234f44b448be363a8b5c4e", + "b6f08470c1cf45dd97d657092267efb9", + "e5bc5d3e749c4423b4fb9c3fc2328281", + "9bcfee9c236f480487c6530f1a475b56", + "2100916100ed468eacc8b746a6c9ba0b", + "28717a14ebaa4708be6467202dc1cc72", + "149a12a56acf4ed5835bfb4e79ac9060", + "e7c2e1cc77bf4a9f901453a987f50f0c", + "f8e5417887bb4815be3cf6c764fb18ef", + "6d87a649081d4a6db73dce63cc68fd05", + "e3fa86ff59f7459f9e8adf2414bcd0ba", + "faf9da9399ff43348ead176f92ceff68", + "971fbbb1e8c7475c99c0813eef1dd955", + "98229bc34d8848e386f90b22ea5d0bde", + "5b519c2321234664bec4fd9113022da3", + "20bc786d1ce84edd843ee174d7c11edc", + "24b48bc1221c45acab931780f51ef89f", + "0336c23eb7264b85b908f94a71564d14", + "2126b3fd17364cf8b7c31b90b236300f", + "bd581a4138fe4eb4b1b98b5d13f9c6e4", + "346c8a532cff426d8a815fdd87576481", + "b051524ebc5e4913929a08198316452a", + "10cc0a373a184a1a8557a663db2220e7", + "5a7102022fec4bc49e6d31bfc1474c24", + "e128c3ccf1324bd5b6bc05b71aad92d2", + "249baa0637614779ab3ea15097c1eea2", + "d95ba50071bb434ca0e1fc5608e83d9b", + "433a0f20e7194101bd56bb7a40ac4665", + "31f7c1ea6a5e4b41b7a37749958e172e", + "84a16620f70b4a0f91a90de18f460fd6", + "b8759548331345079689b26f35a79a77", + "7910e302ff6f4cd8946644406e2c99ad", + "c0c3de1115a64039aeaa67ccc64324bb", + "d2fb9021fdac446695c842d2c792bfa2", + "4a760f234dad418189cbf87761a8501d", + "346fcb32c2d5444a9ad4a08ee14914f6", + "0cfbbc11989c4dd7905f7091218759d7", + "bfe3dde7bcf84031813d3f93c02a1779", + "b2f4575b60d448f7b63144134ddf3d90", + "56380672ce2f4754aee7b7645644c0a5", + "89c87caf2e5e4e6db53132ecb99afe48", + "aa8f6c817dd941aa85621988025478cb", + "96d0554b09f0400eb9368c1a00026a10", + "1ff88b228a8e4c88beebb8c527db136d", + "33fab69359e04180bf754bf3e2ffaf1c", + "80b324a993bb458ea798bc61086ab048", + "b17e681239e44fa2a0ef188d3f4ed0b4", + "9f072d8f80334c0bbbfd98fb7667ef8e", + "db7afa36ff33456c841eb9e262bf64e4", + "cbf45afff6d242fabce8f8c85ef356d4", + "af8f3e3993c247beb049f7208212cf71", + "f0453d4dfdc64ff7bdb0b7b469347ece", + "e529b7ea2fc0414991532d82929eedb5", + "10ab5340121f40f0b7664651ea356c57", + "ba77ceaa78c549238450c116edbde5a6", + "b9e1c8b7407143b89930decf94b4a0cf", + "4ef24dd03bf4481993c955959b436c7a", + "3c5a3337721941078c409b79f532ee21", + "8611029842bb498496be51f074148620", + "18307a79e6934c979ff89cd62c6735f7", + "d236e7740d6444598ca8fbc17a921aa5", + "746788a3733845539f86eccd5435af7a", + "a93ded73f7324dcd8c35c5c941b8d376", + "c66a50a7121947a2aad757f17d02e50d", + "fd418721d3434d26b318a297d4f8a231", + "64ff8bba2d6a42268d685e98dad5d3ba", + "3b9094e941db4acabb47004acff6d5be", + "13879c1049c6437590de3c745eb66c10", + "7e30d8c1c6b9477b85561ff45552f183", + "da34458aa1cd451aa82bd0f1a201054e", + "1e91cd8fd2a942a78f16e64e43647a68", + "86e251461d0641cb816a6536a94d76b9", + "fe80e8d140cc41548750d8d0c033a745", + "31ba3c5b710f44a19a532311b6617e47", + "dfccd141316d446db864f9cee87d5134", + "4996d3e41ade4d1aa77b8584cbd8cd90", + "139887d421954ee08aca9e67f0ae1789", + "767963f09cf1483fb94a3384e6c68393", + "55a1b96224d74308b4c5d3cfc54e6666", + "d7fd69d85bd94c1b9101de88917a56d0", + "e9fe56efd05243c78bf3def4059cc5d5", + "d402754d2c214f129008730d035dfb3e", + "edb0d48ba263454297e61976eb78669b", + "3f0843e7b6f64d4bb6f1787f27f2b5b3", + "9ae9ac7e39294492b02429732a76200a", + "c72181fa23a9422298c8955dd7be1f94", + "16abbacb22d44575a9141cbcb14b6fad", + "bc9c008333e34c8fa5cda6067e91e306", + "c91a2fb1f14244aeb9da258ccd0636ce", + "84372dd446174ad8803d1cd7698c47e3", + "d5ac8b79cdb74d11ae1cea4a253a7813", + "48171f3cdbc945689fa3c99a88efc262", + "a74b89ef8f4c4fe4b92160fa46a3eeb3", + "c96fdd2be2a84390b96eb2b20fbf2231", + "aa15c5e6a8ac4d05862ce73125d7c40f", + "b8ab4b12a6fb4887bb6c0333cbd91ea8", + "0b6b4d4182eb46dcb9bd0a5eca5f60ee", + "c8f02de241004227ba590a1432ea6e84", + "70e78c654d484aa9ab406af8ba91ab0d", + "b918dc885abb450dba336c502b97b861", + "932896d6d10444d4a159306cb31d8d6a", + "93a1ce256ee440ed9084902937a5621d", + "d0f3d668b5dc40dda90b6c845c2ef7b4", + "1650dfc316274bce8ecf6c071c687f1e", + "d928f6cb5fc943129ae571a421b7a2d4", + "7881a6e94d8c4dc8bfe375bedc5eafe7", + "eb3abb7ba6d64b94b65f27177c46b411", + "f3acb74c2c594020aedf6330ee1f1847", + "b83c083a5acf458a9f4ed93f5037b84b", + "615e50faa3e34a3380f2012a65fa3825", + "4a652972f8084b9caaf798d2f99ca6b7", + "a611df0fed2242d586dea0adb1197b44", + "e8336e9cea94474e9874959ca69d81f7", + "802b8b73a4074d5691ed180a879cbdcc", + "cbee27259204437d873d80ab8b70932f", + "a19ec4ba40fd46548f12c4a5b11fbfdc", + "d2abb6d0e55a4b9ab027964c19291751", + "a738f491166d4e90820c6f88f2038cf6", + "d5b57aea91f84665bbc41dbb4dda3230", + "4d62d0426c3649fab4c374d0fe33dbea", + "5351971c88e3442fba201147a6303fa2", + "6ecdc158d0b442e0bb46ddc3909c5039", + "c13b289614424c148473c823e42b854a", + "a0977fe1e5cf409fb3513fa65a83da47", + "dd049390fe7e49f0849e5cf1de250b21", + "65a93ea8237c458dab18cd8fcc3783c1", + "d4e13efd20be41a982c0eab2bc1e50d7", + "cd4f229665d047f6a49ad9b2f7154502", + "d946f3a09f9b4f659a85368d25145aa6", + "546b70c09f344f1f909c1e12e2c4e51b", + "0487417b6f1b4b7e8ebba263c895bf0b", + "f3749087986a4a548f59131c1c74df6f", + "03e2b8789387474d92d449a1946dfceb", + "a96fd2023e2d45848a631e22fb05e028" + ] + }, + "id": "KJeg9zLuGUcS", + "outputId": "0cb2ce90-2cc6-4946-d1be-d377ddba0266" + }, + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "ALIGNMENT BANK: 5-Expert CaptionBERT-8192\n", + "=================================================================\n", + " Device: cuda\n", + " Experts: 5\n", + " Anchors: 512\n", + " Bank dim: 128\n", + "\n", + "=================================================================\n", + "PHASE 0: EXPERT EMBEDDINGS\n", + "=================================================================\n", + " Loading 500,000 captions...\n", + " Got 500,000 captions\n", + "\n", + " Extracting: bert (google-bert/bert-base-uncased, max_len=512)...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/199 [00:00= 3:\n", + " bank_sections.add(f\"bank.{parts[1]}\")\n", + " else:\n", + " bank_sections.add(k)\n", + "for s in sorted(bank_sections):\n", + " keys_in = [k for k in bank_keys if k.startswith(s)]\n", + " params_in = sum(fused[k].numel() for k in keys_in)\n", + " print(f\" {s}: {len(keys_in)} tensors, {params_in:,} params\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# 4. SAVE FUSED model.safetensors\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "sf_path = \"/tmp/model.safetensors\"\n", + "safetensors_save(fused, sf_path)\n", + "size_mb = os.path.getsize(sf_path) / 1e6\n", + "print(f\"\\n✓ model.safetensors ({size_mb:.1f} MB)\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# 5. VERIFY LOAD\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "print(\"\\n Verifying fused model loads...\")\n", + "import sys\n", + "# Use the updated modeling code\n", + "modeling_path = None\n", + "for p in [\"/mnt/user-data/outputs/modeling_caption_bert.py\",\n", + " \"modeling_caption_bert.py\"]:\n", + " if os.path.exists(p):\n", + " modeling_path = p\n", + " break\n", + "\n", + "if modeling_path:\n", + " sys.path.insert(0, os.path.dirname(os.path.abspath(modeling_path)))\n", + " from modeling_caption_bert import CaptionBertConfig, CaptionBertModel\n", + " from safetensors.torch import load_file\n", + "\n", + " config_path = None\n", + " for p in [\"/mnt/user-data/outputs/config_captionbert.json\",\n", + " \"config_captionbert.json\"]:\n", + " if os.path.exists(p):\n", + " config_path = p\n", + " break\n", + "\n", + " with open(config_path) as f:\n", + " cfg_dict = json.load(f)\n", + "\n", + " # Remove non-config keys\n", + " for k in [\"auto_map\", \"architectures\", \"tokenizer_class\",\n", + " \"torch_dtype\", \"transformers_version\",\n", + " \"consensus_models\", \"consensus_alignment\",\n", + " \"consensus_equidistance\", \"training_data\", \"training_samples\"]:\n", + " cfg_dict.pop(k, None)\n", + "\n", + " config = CaptionBertConfig(**cfg_dict)\n", + " model = CaptionBertModel(config)\n", + "\n", + " state = load_file(sf_path)\n", + " missing, unexpected = model.load_state_dict(state, strict=False)\n", + " print(f\" Missing: {len(missing)} {missing[:3] if missing else '[]'}\")\n", + " print(f\" Unexpected: {len(unexpected)} {unexpected[:3] if unexpected else '[]'}\")\n", + "\n", + " # Test forward\n", + " tok = AutoTokenizer.from_pretrained(\"google-bert/bert-base-uncased\")\n", + " inputs = tok([\"A cat on a windowsill\", \"A dog on the beach\"],\n", + " max_length=128, padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\")\n", + " with torch.no_grad():\n", + " out = model(**inputs)\n", + "\n", + " print(f\" last_hidden_state: {out.last_hidden_state.shape}\")\n", + " print(f\" enriched: {out.enriched.shape if out.enriched is not None else 'None'}\")\n", + " print(f\" token_embeddings: {out.token_embeddings.shape}\")\n", + " if out.geometric_context:\n", + " print(f\" geometric_context: {list(out.geometric_context.keys())}\")\n", + "\n", + " cos = (out.last_hidden_state[0] @ out.last_hidden_state[1]).item()\n", + " print(f\" cat↔dog cosine: {cos:.3f}\")\n", + "\n", + " if out.enriched is not None:\n", + " print(f\" enriched dim: {out.enriched.shape[1]} (768 embed + {out.enriched.shape[1]-768} bank)\")\n", + "\n", + " print(\" ✓ Verification passed\")\n", + " del model\n", + "else:\n", + " print(\" ⚠ Could not find modeling_caption_bert.py for verification\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# 6. UPLOAD EVERYTHING\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "print(f\"\\n{'='*65}\")\n", + "print(f\"UPLOADING TO {REPO_ID}\")\n", + "print(f\"{'='*65}\")\n", + "\n", + "# 6a. model.safetensors (fused encoder + bank)\n", + "api.upload_file(path_or_fileobj=sf_path,\n", + " path_in_repo=\"model.safetensors\", repo_id=REPO_ID)\n", + "print(f\"✓ model.safetensors ({size_mb:.1f} MB)\")\n", + "\n", + "# 6b. Bank standalone checkpoint\n", + "if bank_path:\n", + " api.upload_file(path_or_fileobj=bank_path,\n", + " path_in_repo=\"checkpoints/alignment_bank_best.pt\",\n", + " repo_id=REPO_ID)\n", + " print(f\"✓ checkpoints/alignment_bank_best.pt\")\n", + " # Also upload final if exists\n", + " if os.path.exists(\"alignment_bank_final.pt\"):\n", + " api.upload_file(path_or_fileobj=\"alignment_bank_final.pt\",\n", + " path_in_repo=\"checkpoints/alignment_bank_final.pt\",\n", + " repo_id=REPO_ID)\n", + " print(\"✓ checkpoints/alignment_bank_final.pt\")\n", + "\n", + "# 6e. Encoder standalone (for continued training without bank)\n", + "api.upload_file(path_or_fileobj=encoder_path,\n", + " path_in_repo=\"checkpoints/encoder_best.pt\",\n", + " repo_id=REPO_ID)\n", + "print(\"✓ checkpoints/encoder_best.pt\")\n", + "\n", + "# 6f. 500K cache\n", + "print(\"\\n Uploading consensus_500k cache...\")\n", + "cache_files = {\n", + " \"bert.pt\": \"cache_consensus_500k/bert.pt\",\n", + " \"modern.pt\": \"cache_consensus_500k/modern.pt\",\n", + " \"roberta.pt\": \"cache_consensus_500k/roberta.pt\",\n", + " \"albert.pt\": \"cache_consensus_500k/albert.pt\",\n", + " \"distil.pt\": \"cache_consensus_500k/distil.pt\",\n", + " \"captions.json\": \"cache_consensus_500k/captions.json\",\n", + "}\n", + "for local_name, repo_path in cache_files.items():\n", + " local_path = os.path.join(CACHE_DIR, local_name)\n", + " if os.path.exists(local_path):\n", + " size = os.path.getsize(local_path) / 1e6\n", + " api.upload_file(path_or_fileobj=local_path,\n", + " path_in_repo=repo_path, repo_id=REPO_ID)\n", + " print(f\" ✓ {repo_path} ({size:.1f} MB)\")\n", + " else:\n", + " print(f\" ⚠ {local_path} not found, skipping\")\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# 7. VERIFY REPO\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "print(f\"\\n{'='*65}\")\n", + "print(f\"REPO CONTENTS\")\n", + "print(f\"{'='*65}\")\n", + "info = api.model_info(REPO_ID)\n", + "for s in sorted(info.siblings, key=lambda x: x.rfilename):\n", + " size = f\"({s.size/1e6:.1f} MB)\" if s.size and s.size > 100000 else \"\"\n", + " print(f\" {s.rfilename} {size}\")\n", + "\n", + "print(f\"\\nhttps://huggingface.co/{REPO_ID}\")\n", + "print(f\"\\nUsage:\")\n", + "print(f' model = AutoModel.from_pretrained(\"{REPO_ID}\", trust_remote_code=True)')\n", + "print(f' tokenizer = AutoTokenizer.from_pretrained(\"{REPO_ID}\", trust_remote_code=True)')\n", + "print(f\" out = model(**tokenizer('A cat', return_tensors='pt', padding=True))\")\n", + "print(f\" embedding = out.last_hidden_state # (1, 768)\")\n", + "print(f\" enriched = out.enriched # (1, 896)\")\n", + "print(f\" geo = out.geometric_context # dict\")" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "844ca279a1214e9aaf01e0ce607f597b", + "6f81df77f8754e4984edee91b95c06f6", + "5c2eccc71ebd4a419dc32698b1760fd0", + "45e4090dafd94747bd871805029e608e", + "cb0a41a2ef4f4cdfa6d1d1a7f924002c", + "8cb6b1dfd31146419ff33c919f2901cc", + "9fb532dda749453c89c8ab044ad31529", + "78d12e22baad475f947fcda09b5278b8", + "a04c0b91b85945bb834baf72f4183b6c", + "d240e19d83f5499c8694e809fff45fac", + "2d630d5683194818bf6cdfbc927c780b", + "daabbc1ccb994d4c91e49278ad8fc7e9", + "abdd527ee4294e6f9c6fa656f56105cc", + "ddba119e49ae4e4ba80a19d42695652c", + "3628a737a8154beeb0acb3793a396dfb", + "0ad5fae36ddc4977afcf00ac6be6d468", + "63a7abab53aa4f489f0bda1babc0d408", + "dc9d9cbf8c27457f863c5fd43c97492a", + "f831e6630b7147528886e12888898e67", + "772345ef17ea42728007a6edc31f418e", + "630e9a7c293d41bfa126a101ce0cf63f", + "eb079b0bb2224406baf26f0d63e9fcb0", + "f9b8c0e168e04a1b91aac96442e45c4b", + "d7d5ac10cb5045c387abf2831e52d066", + "746c4209461a44308fae532dc1fe3718", + "9f667e0f66df420b906359b35aba546e", + "909ee967105846e8845df4d541951ae2", + "0fa661e22316455aae493914ec79951d", + "79ed5fc5b4264a1a97f472045c7da4b1", + "aa3af48f96c94cc0909e2a4cf2c81dca", + "ec9757b004e24e599d8cea252eb514a5", + "ba2a9dad912746b1bc068c65253f9516", + "dd1fd477b0584b9895207e6764f2aad7", + "9a5bcab81f4f4e6d9d0cf0dc0ad7f072", + "f2844a1d8cd44bb5a319637dfc22bdf0", + "5448afee2fb246f89f6cf11bd5dccbc8", + "73427526fc7d4af4aae82a12183f34fb", + "9a343c75782a415580509fdf2d56b581", + "02998e1a96de4f92a5b6a66ae3d73c14", + "4e587115d78f4964b0f380ab3609f072", + "1b75e405ffc04d3499b5d2767e599ae8", + "1d03959538e3448da85423c192df8e02", + "9967f0d360954a88b38a5ebe4e1bc450", + "3071e444c3694300858adcc023949187", + "f4f687f2230b42a487b76a0c053d7d3c", + "7b0a7b4dacbd431ea1d9ede26840efd5", + "f2ab7b1fd39c40258f361be283a1be71", + "65c0717a4c264140a8f519f9a739dabb", + "5ddd5a7845f74cb2a215e8d00894cd43", + "92f57381b25d4c01b573db158a49a21c", + "28da595390514fff8f3128b6d7b8788f", + "03f4a1bb4ffe4c7a86db79d1a28f68a9", + "57022bfea9a645a5a0081f6c1030c19d", + "68d4e310cf56489e94bd7ec9df3697ce", + "2dc73c08adbc42c1959c7d5a8e5ae6e2", + "be0e1a5f9c6c4209b8aa90a184d271dc", + "540cdd218e2e40a0b8860ab4d23d75e7", + "abd5346d99014ddca5d9a9b862e6c171", + "de3186bec27e453d870688404cb7569f", + "993e8da4216b438da776edb360ffe123", + "330886fd8b7741a49e67c2589dc0d7cc", + "781271b67f134a288cea9ed362373788", + "3a4d04ca69374a81b35dc371005945cb", + "3e26f6f61a134f73ade5d754344f78ba", + "58b0317a4b724a96a67e200820aea5b4", + "8821c2d84a4f4eb4afd3d886b60d1d29", + "aafdfd194a344d46805476fe892af3a1", + "531ab6c2d71145e8adc057385f7817bc", + "a9bff173bdac4e24ad18922d0ba619d5", + "247c1842cd734846bfc682156a6d7156", + "dae98cdc7e644e678d56705284f33367", + "df405bcdfc384f7d910ab7d5062e0999", + "a5ae260280294bf8be1de758ef7e1149", + "79ec96f0a7914deba06ac738d4c30600", + "0675c64897d14cf4afe5f655cefb27ca", + "2c37bec7a7104904bca4bdf312f29da0", + "8f45e2954ee54f0e9fad63f70a85667d", + "06aa2181f5b14f91b7e1c916564c0f1b", + "9b619359df054bcc9cac920733bac1f2", + "278a85927a56473f9eb31fd950522917", + "44ac24f7a99045489434c8961453d280", + "8730284d606a478492c00fa762a745a3", + "3e5d8372c100477294f1155c7e55a08e", + "5750e73b7271438cbed9e6a413641e88", + "bc77094115744d97b78821775ec18f56", + "fdcbc4447f7b42388dad192c54627b58", + "d581bdec81734d68bd5dd7af74f345ff", + "819362f51b7745328e8a3b3be5ad1290", + "aec73061149c42f8a8f9ecd1f5040657", + "a3c2ddcb69dd40bf9289e090d8c235cd", + "0a160f09ddfb40bf983fda72038dcff1", + "034640b2d7bf450c837f3b2a996be2ff", + "f429d9eedc8f4495b11534533fa7d1a4", + "ff95002595e84e8a805ec6f22fd34677", + "d43594d8e7884c4ab2b7ccac1f1c0227", + "72ccbf31bee9462283e8ac9aad8d529f", + "d954c2064cba42f8b561a6f067fbf53b", + "36c6a171fa4c4dfc87a121243d90cef4", + "2bff46e20c084c4798dd9012950d6bd5", + "3ea7d1536b1a487dbeb3200dca65f573", + "a4b28207be614284b4da30bd1c243ab7", + "cc42746c36b04efe9d0019516ab15b24", + "89d355da1e4245d9b8c73f58f45e16ac", + "475c3ee99a814b41befb04df1a0dcb1b", + "47deda8ea4464f9fb573ce607ec0b457", + "5b3fc4dd048a44cc802de7d36a554840", + "9ba7853e89064fc0baf96032bef06ea7", + "70536b254fec4142b3e04a9b8598836f", + "9db2c00ace564417b68231cf51486282", + "9e85c55583a9464992c6eca4bdf90b91", + "b952d107e80d4ddeb9f61397e3968659", + "06b2f006b1ab48359a36b91a21b48e87", + "3983fbc28cf8447e872e9495714905e1", + "a1c68a4910944d4cbaa582d6621497a3", + "040ea1a742df41d1b8ba7754bf93230d", + "c76fee1c9ae74473b2aa32ae8a030de6", + "afd6158d26e1491ca4cffcbb8cab0c8f", + "6d08deefc3fe43208ecdd551cdd670a1", + "30d525a65572435292c59fa8c60347d4", + "71c8482e69594de7841308c4afb8ddb6", + "0a4e2574d38d4f3eb84297c4ae99cf93", + "5570f9abda9747d8b81e459efc6f420d", + "ac7892cc541e4a858db94a82b5ffdf46", + "3954a8f50e8348419301305de7637203", + "36a1da91b520487cb00bb3ae9f68322a", + "db412f7b422f443ab9fc02f5c8a7267a", + "5852e1611ff646b587dcd59d2c2cd7fa", + "63af04e265734e4ea0edbabc3637ac77", + "722d6f4a94ab4270b4bd8438e1a1b2d3", + "bc4f89591ebc47298d8d1732dbdc70ef", + "f8f2092ad53649019835aec315d2ad6d", + "8e2755a19c3f418fa777030efd951486", + "b84fd89752544a1bab4f039217c52d83", + "f04c917e1ffa4918ad00ecbca177741b", + "0bb086c92ef7429f9ab48c89e4f847fc", + "57b0ec87b7ae41bdaf3cc565c6612de8", + "9428c5c782384f139b301d0e4146c475", + "249e7f3721c74243bf59cdb3b65f2350", + "e320cb11d8cc44638fc997e4bc3066b6", + "1b0b0e4642b5444eb7188f934a1a45a9", + "7ae63114504d421283e7bd5df8fe5c2d", + "b20ca3a889f249c5a4ec61faab39f3b9", + "00b6693fedc24dd4b6018b1d7ad66a22", + "7f33873d36b045ddbf4cebab46902c8b", + "142260d1a68e419981476cf344fede3d", + "01a38c16c788493ca07e128f96afae89", + "0f8126ee7e9a4740a4f084ddc716a1dd", + "3fb95746b95540178f0e3037429fc60c", + "28a5c117317c43388b93b8550b78d4bf", + "1210f005a3b94c9ea34b198213a809be", + "e056dfcd0ed64eacb4b4bafa5acf5ac8", + "5ab6f5b71c1441509fe834f0bf42cc58", + "da55e960029542bf97e9b6190471f818", + "9e4dbbdf7d694ce491c2bad345e437be", + "7fdb7fdcc62c497e86fb36684ce5a1db", + "e6899428acb04fd2b3271a1a6c07c8fe", + "8b5f6d2bd3fd4712a1cd7e3c00a42cf7", + "a15fad0679f54591ab0c7a796da53544", + "33bedf27e64749689e39943342c9bacf", + "03c7da3ceb214befb2419f3bb41d12c7", + "37eaf99eac1a4424821bb674fb9fdd9f", + "bd788d18620e4310a768c5ecc80f942b", + "8bf1b010a1da445d90ca090769ff70d7", + "56d38e95935744a59eafe524d4ad2b31", + "f3b28548337f4908b469735773ef341d", + "a4b3c3215e5d45bcbd6ab93d54264695", + "0d2a88f02744476586ec8844867841c7", + "9b78b5350eff443f936606c5cd447b81", + "80be16cfb01d49dba2fc9b1cd8f46a81", + "306fe1a1ed994c44ab86cb3991d7692d", + "9b1eecbf4ef6497f976daed1aac23d20", + "8e5eeb51a5c04f119f60758128405426", + "45aec9e663db4c99a402ce5292ca8587", + "623ac375257342acb2224f4d540f0ddb", + "e150efa1f71443cd86716057c82c1b85", + "04bdecdb857345d1aa8e7fe072e58e7a", + "d39027271be948c4ac296a71d3562170", + "d88e715865f848ee8171f00e1980a99e", + "afed383eb0e740e1a4bca5fd2fd52bb0", + "962f8493ad9c4bb0a59b6f9cd25d03cc", + "4b0b4345ad9c4c098cbbce36189c1153", + "84b328f84fd242bc9487a17aa3cf1b0f", + "0323fa19808b42f9880d3c6e44c30b13", + "3c931b55244c4027bc6e18fe86511ec8", + "f3f356ad4085424584eefca058a79861", + "a89bdca1241846e3a05eae708fefbc81", + "6d6a8ba5f6af4d379f4838e327826e4f", + "459d7be6b5644d12a407e8960209a4d5", + "9b7530dd208249adb88db7b3ae05ba18", + "9d6061b361a14d28a50bf5fc8eb28c23", + "cfa4a32985ff4be98565f980ef2eab18", + "cc2b2c6a6341499fbf82c949bc275cbe", + "d5a58ce74771424780fe9b6fcd8538a9", + "0fa3ebd0d81f4444970a9ddfd278e3cd", + "3fd9968c71274aa581390b1fed4d66f9", + "644cfc46d21140e1a9e1f96f418e5463", + "e67bd46190674559aca94e0321f7e4ae", + "6d37220964ac4d3c9f9429ffbd186377", + "3aabe6114a6442c2bde35b5ae6ca5e58", + "6c5b50a3caa1424ebbdd373f2dab19a4", + "fe0167cd18664b85ab6b0b10353ad6cd", + "aa1c6b16ec0145e0a56ddf970ef90246", + "ade78fd5481f4661bba0f50f15fcaaa5", + "488a0b550fa44a32a2b3c1cb0d137a77", + "71de7c2dce1b41af91ff969ba9e60afe", + "e799c7d3888641f4a803b00e46c63f3b", + "f0ef1e54e7fc487592b15533a8733a73", + "133ac47e177a4bb1ad94a5c5c81b9708", + "c635d8b0865f4a408ec1bf312d29e8c6", + "b5992e52d30e42018359dd14a88de423", + "72ae7f43086a4f77b0da51ab87ffddb9", + "f3b16ac9b00d49378c2f0b6f4583d53a", + "fad4f7d6a9af4e5e9b31c786b211a512", + "5d9d457eb7804ae79da4241af0116884", + "e2b774bddb22445d904b1b6b8403fa27", + "4e5dd175914445709907787c0f62a0c9", + "547843a2d89d4151af1a8da52c927dc3", + "f00dcfac18d343f28dedc375f65bcf0f", + "e06f304b6495429e937afae8875dacc0", + "d30624f8745e4996971c52f42bd1e433", + "b105233668c6490b9e8bb1cd2deec55f", + "92284c99f03b4de6b7cb93b87651f71e", + "cf16652f4a0b441685515f512df7bd05", + "4f97a0d0185e4a47981944df05ea0722", + "e9fdc759f3174970ba20d86ea713890c", + "db9bec5b58ab4189954f73ec45c0234d", + "7c0f34b3d6fa41dfbe25ecb354000243", + "a80d180db47b404d82167e86a4bab129", + "ddbd666dfed3416abfc8f0d26be9dadc", + "bbe45d6dbd034ba391648511d4b8761e", + "244d94f8d35840038411f865f5b3f48f", + "d4ab33695163401090a9364f9701a670", + "97cf8215706149518fc00ba3f720bbfe", + "052dd3ef5d074e3da44fedc407ac8926", + "1bfdbf63cfa549ce8d844a9c919cde8a", + "ced0810573d44581bb8bf4902c84d4cb", + "9726b22ce55948859094e118438521e7", + "76a2507027674d0d935ff698064e9fd0", + "dbdb8c81c34b4a18b624053cfd857ab1", + "43980f5d4e15469cab101b804bf8321e", + "118dc7ab862d4c668373c885a0deb53f", + "18104f6c84324d208c8e6d6476a3ab11", + "01ae48f5c8eb4961a459010fb763375f", + "9dbaf26407b2429c957a26882371558e", + "0d298d4879a747bfbf62e69f9a3a73de", + "99945a38c13f424cabe8b8edb6d9bc9c", + "cbdc97ec3cac4aadbf105a52d2c95975", + "a3bb128d75404b7f84539f1ccc4253cb", + "05c165f45f1641a58161ca09ff58b1e2", + "0f6619c22c494bb6ba535acc8a56764d", + "7258a7ef572c4c7a815d255acf887a22", + "b2228d50851a41e8b50739c0bf01a01d", + "a128d1ab378c4de5951f6efb58d8a9e8", + "64826e72527f403592c5e1cd02db3cc4", + "96892a8638624211a0ebb6edbac9b291", + "3c3fc5ef68944753b136932dfb295554", + "5b7fb3754b1c4f23bdc92acb0ce3406c", + "4afc86dddc0543efb4f15f9544ae2ef7", + "2a97d7a2107b488488a03d349c71f6e6", + "8ece2270cb9c4c3d9591b6340d681519", + "c3b2417a4cd14d5eb3463a58f7e9a5fb", + "7e7c7a946a934ec3951c39fe2c8b600d", + "121419d72ac9427a9fe7c29e5c7bf035", + "96f8af9f08504044bb13345f63f22b3d", + "8abef333421b446d89db434f3cf44673", + "4237d6f0583c472bb964a08b9bfd9f86", + "7c8646f25af140888b95fa91abcb9ec8", + "2740216ce3f0424f8077e11e228652fd", + "66bf0f206ce44a08b7501972c624c7cd", + "781aa0ac2f8c4e11b06fd9a864047333", + "920f1e10313048ae8db5dd32059b7f55", + "598607abfbfa4afeac842c8f63aabe99", + "4a7e8954f44e4cea85b683a38d377f56", + "510e6a8a5ea541e79353561a6e026de7", + "37328c6ca09a4743bf7ea0a0a15aa757", + "452e946cb8b64b549b034f09a477425e", + "0b9ab15129d5494b9c1742a44f2d9229", + "abea04e26d2e401d86e8cbd27b36e57b", + "72355f30c970406598ea0edde4ed5a1d", + "0b10bd468f5041bfbcb41221a9695e4e", + "1a1846fc50734c538007354f8d1aeecc", + "12650ad15d5049c5ae2b5edac0f3e098", + "687a356cad924ed8b5ec48e3e9dcb02f", + "5b888c3082d94c31bf46ef5cc83730ff", + "ed65d50863b64a529bc49b33b4592b41", + "e597ee9e7ffc497ca366143554a23192", + "3d76035339604e52aaabb287540be49a", + "25ff659b9fe14dd4b5b0c7addce38cdc", + "d4b8b1bdcbd14085a87540f6ec28f470", + "2b6f4a6f1c4f4f24a1ba668e89e2cc00", + "1356858c469e457fa37f43e675c3e6b0", + "11980b18a65748c788cd43401737b53c", + "a6c15f566fc84f7e90cb3083aef8194b", + "42b65b6356b74c959751141b8922fceb", + "5dc1f895e77b42018dcfe3fbffa8567b", + "b48da0a0dcbd40ccb83acf615c754cf7", + "92ea4373e26d4a6da15ad0f5fa665c45", + "2a10ca6288854ea4be8ddfc0a112d1ba", + "2991907614494da98e364e7b4b261668", + "3c175404310943b2b975ab724d32c2a7", + "1e063330ec964419a772d37e39adfb26", + "0f01672fccbf42cf91bb3aca938d1e17", + "5ca32ad1e18143b599789ebad84c0ee9", + "b776d149237d46ef85acd2c55db891ab", + "cae01e43ef9548efa678ed58aef7048c", + "f05c0f0910384ec8830c1d436f7a883c", + "964a1b99e204423293abc2233390b5ec", + "5f3d5b2cb2df45ab92ca01d28e02c04e", + "d604e7df44044904816c90d442134d32", + "d3ce02cd5815453d90f60952bb8a8fce", + "85bd6b735db446dca5e92aef832d66f8", + "bdd138b348024ba19efc9d176a90950a", + "7e83b5c2514142fda892d65abbf3adea", + "cade1456401f42b298a553d4121da01d", + "7f5678a3a79d483bad005f8ed43a4c76", + "94fea6d51fe1469d92f5283a42639f30", + "e9d4c8fbc7464ea490289252403276c9", + "0211210cc1fc469da30984bd88c9c572", + "63937d91a06f43e8b6d152b6198def11", + "fa30381c3e2a48c6a2d2738173667d63", + "6edd557d3a634020999b0c1a47c44069", + "3fdad0148a8c4df187203d1d03dca0ef", + "43df18789778424da8a3fbaf29c33715", + "65bb38bd08cd449b968dea79f08fd027", + "796a0b14aa4144fe94ca38f09620e5f1", + "3e477950d41e4d689207fe4de3da3aa8", + "28d93f9b274e428782b6be59665f9984", + "8dcebf5b9b0043b09f3eae87e066798d", + "95f6f2f2332d4537860fe746b8a4bd47", + "4870ff7175c34edbaa2f20fe54d2929a" + ] + }, + "id": "go985XH3gse6", + "outputId": "d41eda8a-6436-4ab8-b25a-9d81d0e2b419" + }, + "execution_count": 5, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "UPLOAD: AbstractPhil/geolip-captionbert-8192\n", + "=================================================================\n", + " Downloading encoder weights from HF...\n", + " Encoder: 82 tensors from /root/.cache/huggingface/hub/models--AbstractPhil--geolip-captionbert-8192/snapshots/8dbe9e65befe677a679e9b4b47e743d88d5e6e1a/best_model.pt\n", + " Encoder params: 25,958,016\n", + " Bank: 30 tensors from alignment_bank_best.pt\n", + " Bank params: 6,466,999\n", + "\n", + " Fused state dict: 112 tensors, 32,425,015 params\n", + " Encoder keys: 82\n", + " Bank keys: 30\n", + " bank.anchors: 1 tensors, 393,216 params\n", + " bank.expert_means: 5 tensors, 3,840 params\n", + " bank.expert_rotations: 5 tensors, 2,949,120 params\n", + " bank.expert_whiteners: 5 tensors, 2,949,120 params\n", + " bank.geo_proj: 8 tensors, 171,648 params\n", + " bank.target_cross_cos_mean: 1 tensors, 1 params\n", + " bank.target_cross_cos_std: 1 tensors, 1 params\n", + " bank.target_cv: 1 tensors, 1 params\n", + " bank.target_disagreement_ratio: 1 tensors, 1 params\n", + " bank.target_mean_cos: 1 tensors, 1 params\n", + " bank.target_spectral: 1 tensors, 50 params\n", + "\n", + "✓ model.safetensors (129.7 MB)\n", + "\n", + " Verifying fused model loads...\n", + " ⚠ Could not find modeling_caption_bert.py for verification\n", + "\n", + "=================================================================\n", + "UPLOADING TO AbstractPhil/geolip-captionbert-8192\n", + "=================================================================\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "844ca279a1214e9aaf01e0ce607f597b" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "daabbc1ccb994d4c91e49278ad8fc7e9" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " /tmp/model.safetensors : 74%|#######3 | 96.0MB / 130MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "f9b8c0e168e04a1b91aac96442e45c4b" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "No files have been modified since last commit. Skipping to prevent empty commit.\n", + "WARNING:huggingface_hub.hf_api:No files have been modified since last commit. Skipping to prevent empty commit.\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "✓ model.safetensors (129.7 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "9a5bcab81f4f4e6d9d0cf0dc0ad7f072" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "f4f687f2230b42a487b76a0c053d7d3c" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " alignment_bank_best.pt : 97%|#########7| 25.1MB / 25.9MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "be0e1a5f9c6c4209b8aa90a184d271dc" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "✓ checkpoints/alignment_bank_best.pt\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "aafdfd194a344d46805476fe892af3a1" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "06aa2181f5b14f91b7e1c916564c0f1b" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " alignment_bank_final.pt : 4%|4 | 1.15MB / 25.9MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "aec73061149c42f8a8f9ecd1f5040657" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "✓ checkpoints/alignment_bank_final.pt\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "3ea7d1536b1a487dbeb3200dca65f573" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "b952d107e80d4ddeb9f61397e3968659" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...3d88d5e6e1a/best_model.pt: 92%|#########2| 95.9MB / 104MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "5570f9abda9747d8b81e459efc6f420d" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "✓ checkpoints/encoder_best.pt\n", + "\n", + " Uploading consensus_500k cache...\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "b84fd89752544a1bab4f039217c52d83" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "7f33873d36b045ddbf4cebab46902c8b" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...de/consensus_500k/bert.pt: 0%| | 556kB / 1.54GB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "7fdb7fdcc62c497e86fb36684ce5a1db" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ cache_consensus_500k/bert.pt (1536.0 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "a4b3c3215e5d45bcbd6ab93d54264695" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "d39027271be948c4ac296a71d3562170" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " .../consensus_500k/modern.pt: 0%| | 3.86MB / 1.54GB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "459d7be6b5644d12a407e8960209a4d5" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ cache_consensus_500k/modern.pt (1536.0 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "3aabe6114a6442c2bde35b5ae6ca5e58" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "b5992e52d30e42018359dd14a88de423" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...consensus_500k/roberta.pt: 0%| | 3.85MB / 1.54GB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "b105233668c6490b9e8bb1cd2deec55f" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ cache_consensus_500k/roberta.pt (1536.0 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "d4ab33695163401090a9364f9701a670" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "01ae48f5c8eb4961a459010fb763375f" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " .../consensus_500k/albert.pt: 0%| | 3.88MB / 1.54GB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "64826e72527f403592c5e1cd02db3cc4" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ cache_consensus_500k/albert.pt (1536.0 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "8abef333421b446d89db434f3cf44673" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "452e946cb8b64b549b034f09a477425e" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " .../consensus_500k/distil.pt: 0%| | 3.90MB / 1.54GB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "3d76035339604e52aaabb287540be49a" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ cache_consensus_500k/distil.pt (1536.0 MB)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Processing Files (0 / 0) : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "2a10ca6288854ea4be8ddfc0a112d1ba" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "New Data Upload : | | 0.00B / 0.00B " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "d604e7df44044904816c90d442134d32" + } + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + " ...sensus_500k/captions.json: 3%|2 | 6.73MB / 232MB " + ], + "application/vnd.jupyter.widget-view+json": { + "version_major": 2, + "version_minor": 0, + "model_id": "fa30381c3e2a48c6a2d2738173667d63" + } + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + " ✓ cache_consensus_500k/captions.json (232.4 MB)\n", + "\n", + "=================================================================\n", + "REPO CONTENTS\n", + "=================================================================\n", + " .gitattributes \n", + " README.md \n", + " benchmarks/benchmark_post.py \n", + " benchmarks/captionbert_8192_post_upgrade_60epoch_5test.json \n", + " benchmarks/captionbert_8192_post_upgrade_60epoch_5test_output.txt \n", + " benchmarks/early_bench.py \n", + " best_model.pt \n", + " cache_consensus_500k/albert.pt \n", + " cache_consensus_500k/bert.pt \n", + " cache_consensus_500k/captions.json \n", + " cache_consensus_500k/distil.pt \n", + " cache_consensus_500k/modern.pt \n", + " cache_consensus_500k/roberta.pt \n", + " checkpoints/alignment_bank_best.pt \n", + " checkpoints/alignment_bank_final.pt \n", + " checkpoints/encoder_best.pt \n", + " checkpoints/model_e10.pt \n", + " checkpoints/model_e20.pt \n", + " checkpoints/model_e30.pt \n", + " colab_deep_analysis.py \n", + " colab_test_script.py \n", + " config.json \n", + " final_trains/base_best_model_e30.pt \n", + " final_trains/base_final_model_e60.pt \n", + " model.safetensors \n", + " modeling_caption_bert.py \n", + " tokenizer.json \n", + " tokenizer/tokenizer.json \n", + " tokenizer/tokenizer_config.json \n", + " tokenizer_config.json \n", + " trainers/nil_head_trainer_cross_entropy_fail.py \n", + " trainers/nil_head_trainer_full_geometric_losses_CE_fail.py \n", + " trainers/trainer_alignment_base.py \n", + " trainers/trainer_alignment_base_8192_upgrade.py \n", + " trainers/trainer_anchor_bank_attempt_1.py \n", + " training_metrics/bank_training_try1_500k_output.txt \n", + " training_metrics/metrics.json \n", + "\n", + "https://huggingface.co/AbstractPhil/geolip-captionbert-8192\n", + "\n", + "Usage:\n", + " model = AutoModel.from_pretrained(\"AbstractPhil/geolip-captionbert-8192\", trust_remote_code=True)\n", + " tokenizer = AutoTokenizer.from_pretrained(\"AbstractPhil/geolip-captionbert-8192\", trust_remote_code=True)\n", + " out = model(**tokenizer('A cat', return_tensors='pt', padding=True))\n", + " embedding = out.last_hidden_state # (1, 768)\n", + " enriched = out.enriched # (1, 896)\n", + " geo = out.geometric_context # dict\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "# ============================================================================\n", + "# INFERENCE TEST: CaptionBERT-8192 with Alignment Bank\n", + "# ============================================================================\n", + "\n", + "from transformers import AutoModel, AutoTokenizer\n", + "import torch\n", + "\n", + "REPO_ID = \"AbstractPhil/geolip-captionbert-8192\"\n", + "\n", + "print(\"Loading model...\")\n", + "model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True)\n", + "model.eval()\n", + "print(f\" Parameters: {sum(p.numel() for p in model.parameters()):,}\")\n", + "print(f\" Bank: {'present' if hasattr(model, 'bank') and model.bank is not None else 'MISSING'}\")\n", + "\n", + "print(\"Loading tokenizer...\")\n", + "tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + "print(f\" Vocab: {tokenizer.vocab_size}\")\n", + "\n", + "texts = [\n", + " \"A cat sitting on a windowsill watching birds outside\",\n", + " \"A golden retriever playing fetch on the beach at sunset\",\n", + " \"A still life painting with flowers and fruit on a table\",\n", + " \"An aerial photograph of a city skyline at night\",\n", + " \"A child riding a bicycle through autumn leaves in a park\",\n", + " \"a girl performing an action\",\n", + " \"a boy performing an action\",\n", + " \"a woman performing an action\",\n", + " \"a man performing an action\",\n", + "]\n", + "\n", + "inputs = tokenizer(texts, max_length=8192, padding=True,\n", + " truncation=True, return_tensors=\"pt\")\n", + "\n", + "with torch.no_grad():\n", + " outputs = model(**inputs)\n", + "\n", + "emb = outputs.last_hidden_state\n", + "print(f\"\\n Embedding: {emb.shape}\")\n", + "print(f\" Norms: {[f'{n:.4f}' for n in emb.norm(dim=-1).tolist()]}\")\n", + "\n", + "if outputs.enriched is not None:\n", + " print(f\" Enriched: {outputs.enriched.shape} (768 + {outputs.enriched.shape[1] - 768} bank)\")\n", + "else:\n", + " print(f\" Enriched: None (bank not loaded)\")\n", + "\n", + "print(f\" Tokens: {outputs.token_embeddings.shape}\")\n", + "\n", + "if outputs.geometric_context:\n", + " print(f\"\\n Geometric context:\")\n", + " for k, v in outputs.geometric_context.items():\n", + " print(f\" {k}: {v:.4f}\" if isinstance(v, float) else f\" {k}: {v}\")\n", + "\n", + "print(f\"\\n Pairwise cosine similarity:\")\n", + "sim = emb @ emb.T\n", + "for i in range(len(texts)):\n", + " for j in range(i+1, len(texts)):\n", + " print(f\" [{i}]↔[{j}]: {sim[i,j]:.3f} ({texts[i][:40]}↔{texts[j][:40]})\")\n", + "\n", + "if outputs.enriched is not None:\n", + " enr = outputs.enriched\n", + " enr_sim = torch.cosine_similarity(enr.unsqueeze(1), enr.unsqueeze(0), dim=-1)\n", + " print(f\"\\n Enriched pairwise (768+bank):\")\n", + " for i in range(5):\n", + " for j in range(i+1, 5):\n", + " delta = enr_sim[i,j].item() - sim[i,j].item()\n", + " print(f\" [{i}]↔[{j}]: {enr_sim[i,j]:.3f} (Δ={delta:+.3f} from raw)\")\n", + "\n", + "if hasattr(model, 'encode'):\n", + " e = model.encode([\"Hello world\", \"Testing the encoder\"], tokenizer=tokenizer)\n", + " print(f\"\\n encode() output: {e.shape}\")\n", + " print(f\" encode() cosine: {(e[0] @ e[1]).item():.3f}\")\n", + "\n", + "print(\"\\n✓ All tests passed\")" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "f46f04d5073945e2add5ac5a82747534", + "36d2eeeb7d6d43e2802991a7ee948df8", + "12d840fff521422eade1ce9792fed91f", + "d31a9a5560d84d5eb94f994682f0454a", + "b7077c26d0004b82ae2c8b1065804188", + "5bf5ef0ca2514e5595ed03f2b3d32dd9", + "1d3d8ca8a3854b0a83db4df3c15f9a6d", + "1d63c2f6c3d641b9b041a0e5142ea891", + "bfcda213611c4943859fc5b8b309a31b", + "3ebfa3ed6cd94159bd339766e3a6bb47", + "329e3afa542d4c788695e60cb541d4a4", + "ea3b041d08e7434498a37af80aed30c3", + "ed5026404e97440e948c9207fc80eaa6", + "8a12c418435c4446ae67db9a0285b27b", + "630ab1aa01bb4b26b430a68ab996a35f", + "b7b33f7e5d4b445a958fc936350bdd33", + "288a439304574b978a1a169c934b4f4f", + "b521d31ddc214bc8a73499fb947615d3", + "85c078ae16134042996a13d1585ac24f", + "ae5bd77af032451b8d21687f60c4f177", + "0ea428f6de0d45bd8394be630f8e1d93", + "a1316b6b0692402e9127f9eec611c6a1" + ] + }, + "id": "HHyjutMqjHnn", + "outputId": "3655eb1c-a4bd-4582-aef5-d17a9a352bc7" + }, + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Loading model...\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "A new version of the following files was downloaded from https://huggingface.co/AbstractPhil/geolip-captionbert-8192:\n", + "- modeling_caption_bert.py\n", + ". Make sure to double-check they do not contain any added malicious code. To avoid downloading new versions of the code file, you can pin a revision.\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "model.safetensors: 0%| | 0.00/130M [00:00 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " from transformers import AutoModel, AutoTokenizer\n", + " from datasets import load_dataset\n", + "\n", + " # ── Load model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING MODEL\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + " for p in model.parameters():\n", + " p.requires_grad = False\n", + "\n", + " has_bank = model.bank is not None\n", + " print(f\" Model: {sum(p.numel() for p in model.parameters()):,} params (frozen)\")\n", + " print(f\" Bank: {'present' if has_bank else 'absent'}\")\n", + "\n", + " # ── Load SNLI ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING SNLI\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ds = load_dataset(\"stanfordnlp/snli\")\n", + " train_ds = ds[\"train\"].filter(lambda x: x[\"label\"] >= 0)\n", + " val_ds = ds[\"validation\"].filter(lambda x: x[\"label\"] >= 0)\n", + " print(f\" Train: {len(train_ds):,} Val: {len(val_ds):,}\")\n", + "\n", + " MAX_TRAIN = 549000 # full SNLI\n", + " MAX_VAL = 9800\n", + "\n", + " # ── Pre-encode ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PRE-ENCODING\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " @torch.no_grad()\n", + " def encode_pairs(dataset, max_n, batch_size=256):\n", + " dataset = dataset.select(range(min(max_n, len(dataset))))\n", + " all_p_enr, all_h_enr = [], []\n", + " all_labels = []\n", + "\n", + " for i in tqdm(range(0, len(dataset), batch_size), desc=\" Encoding\"):\n", + " j = min(i + batch_size, len(dataset))\n", + " batch = dataset[i:j]\n", + "\n", + " p_in = tokenizer(batch[\"premise\"], max_length=128,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + "\n", + " h_in = tokenizer(batch[\"hypothesis\"], max_length=128,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " h_out = model(**h_in)\n", + "\n", + " p_feat = p_out.enriched if p_out.enriched is not None else p_out.last_hidden_state\n", + " h_feat = h_out.enriched if h_out.enriched is not None else h_out.last_hidden_state\n", + "\n", + " all_p_enr.append(p_feat.cpu())\n", + " all_h_enr.append(h_feat.cpu())\n", + " all_labels.append(torch.tensor(batch[\"label\"]))\n", + "\n", + " return {\n", + " \"p\": torch.cat(all_p_enr),\n", + " \"h\": torch.cat(all_h_enr),\n", + " \"labels\": torch.cat(all_labels),\n", + " }\n", + "\n", + " train_data = encode_pairs(train_ds, MAX_TRAIN)\n", + " val_data = encode_pairs(val_ds, MAX_VAL)\n", + "\n", + " d_enriched = train_data[\"p\"].shape[1]\n", + " d_raw = 768\n", + " d_bank = d_enriched - d_raw\n", + " print(f\" Enriched: {d_enriched} (raw={d_raw} + bank={d_bank})\")\n", + " print(f\" Train: {train_data['labels'].shape[0]:,} Val: {val_data['labels'].shape[0]:,}\")\n", + "\n", + " for label, name in [(0, \"entailment\"), (1, \"neutral\"), (2, \"contradiction\")]:\n", + " n_l = (train_data[\"labels\"] == label).sum().item()\n", + " print(f\" {name}: {n_l:,} ({n_l/len(train_data['labels']):.1%})\")\n", + "\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # Move to GPU\n", + " train_p = train_data[\"p\"].to(DEVICE)\n", + " train_h = train_data[\"h\"].to(DEVICE)\n", + " train_labels = train_data[\"labels\"].to(DEVICE)\n", + " val_p = val_data[\"p\"].to(DEVICE)\n", + " val_h = val_data[\"h\"].to(DEVICE)\n", + " val_labels = val_data[\"labels\"].to(DEVICE)\n", + "\n", + " # ── Build CompConv NLI ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL CONV NLI HEAD\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli = CompConvNLI(\n", + " d_raw=d_raw, d_bank=max(d_bank, 1),\n", + " d_path=128, n_components=5, n_classes=3, dropout=0.1\n", + " ).to(DEVICE)\n", + " n_head_params = sum(p.numel() for p in nli.parameters())\n", + " print(f\" Head params: {n_head_params:,}\")\n", + "\n", + " # ── Training ──\n", + " EPOCHS = 10\n", + " BATCH = 512\n", + " LR = 5e-4\n", + " n_train = train_labels.shape[0]\n", + " n_batches = n_train // BATCH\n", + "\n", + " optimizer = torch.optim.AdamW(nli.parameters(), lr=LR, weight_decay=0.01)\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * EPOCHS, eta_min=1e-6)\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({EPOCHS} epochs, {n_batches} batches/epoch)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " best_val_acc = 0.0\n", + " for epoch in range(EPOCHS):\n", + " nli.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss, total_correct, n = 0, 0, 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + "\n", + " logits, _ = nli(train_p[idx], train_h[idx])\n", + " labels = train_labels[idx]\n", + " loss = F.cross_entropy(logits, labels)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(nli.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " total_correct += (logits.argmax(-1) == labels).sum().item()\n", + " total_loss += loss.item()\n", + " n += len(idx)\n", + "\n", + " elapsed = time.time() - t0\n", + " train_acc = total_correct / max(n, 1)\n", + " train_loss = total_loss / max(n // BATCH, 1)\n", + "\n", + " # Validation\n", + " nli.eval()\n", + " with torch.no_grad():\n", + " val_n = val_labels.shape[0]\n", + " val_correct = 0\n", + " val_loss_sum = 0\n", + " all_preds, all_labs = [], []\n", + " path_info = None\n", + "\n", + " for i in range(0, val_n, 512):\n", + " j = min(i + 512, val_n)\n", + " logits, info = nli(val_p[i:j], val_h[i:j])\n", + " labs = val_labels[i:j]\n", + " val_correct += (logits.argmax(-1) == labs).sum().item()\n", + " val_loss_sum += F.cross_entropy(logits, labs, reduction=\"sum\").item()\n", + " all_preds.append(logits.argmax(-1).cpu())\n", + " all_labs.append(labs.cpu())\n", + " if path_info is None:\n", + " path_info = info\n", + "\n", + " val_acc = val_correct / val_n\n", + " val_loss = val_loss_sum / val_n\n", + " preds = torch.cat(all_preds)\n", + " labs_all = torch.cat(all_labs)\n", + "\n", + " acc_ent = (preds[labs_all == 0] == 0).float().mean().item() if (labs_all == 0).sum() > 0 else 0\n", + " acc_neu = (preds[labs_all == 1] == 1).float().mean().item() if (labs_all == 1).sum() > 0 else 0\n", + " acc_con = (preds[labs_all == 2] == 2).float().mean().item() if (labs_all == 2).sum() > 0 else 0\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s\")\n", + " print(f\" Task: loss={train_loss:.4f} t_acc={train_acc:.4f} v_acc={val_acc:.4f} v_loss={val_loss:.4f}\")\n", + " print(f\" Per-class: ent={acc_ent:.3f} neu={acc_neu:.3f} con={acc_con:.3f}\")\n", + " if path_info:\n", + " top3 = path_info[\"top_paths\"][:3]\n", + " path_str = \" \".join(f\"{comp}={w:.3f}\" for comp, w in top3)\n", + " print(f\" Paths: {path_str} spread={path_info['weight_spread']:.4f}\")\n", + " print(f\" Protos: sim={path_info.get('proto_spread', 0):.4f} \"\n", + " f\"temp={path_info.get('temperature', 0):.2f}\")\n", + "\n", + " if val_acc > best_val_acc:\n", + " best_val_acc = val_acc\n", + " torch.save(nli.state_dict(), \"nli_conv5d_best.pt\")\n", + " print(f\" ★ New best: {val_acc:.4f}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # PATH ANALYSIS\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PATH WEIGHT ANALYSIS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli.load_state_dict(torch.load(\"nli_conv5d_best.pt\", weights_only=True))\n", + " nli.eval()\n", + "\n", + " weights = F.softmax(nli.path_weights, dim=0).cpu().tolist()\n", + " ranked = sorted(zip(nli.compositions, weights), key=lambda x: -x[1])\n", + " print(f\"\\n {'Path':<25} {'Weight':>8} {'Type':<15}\")\n", + " print(f\" {'-'*50}\")\n", + " for comp, w in ranked:\n", + " if len(comp) == 1:\n", + " ptype = \"holistic\"\n", + " elif all(c == 1 for c in comp):\n", + " ptype = \"independent\"\n", + " elif comp[0] >= 3:\n", + " ptype = \"premise-heavy\"\n", + " else:\n", + " ptype = \"mixed\"\n", + " bar = \"█\" * int(w * 100)\n", + " print(f\" {str(comp):<25} {w:>8.4f} {ptype:<15} {bar}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # COMPOSITIONAL ORDER TEST\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL ORDER TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " label_names = [\"entailment\", \"neutral\", \"contradiction\"]\n", + "\n", + " test_pairs = [\n", + " (\"a potato on top of a table\", \"a table on top of a potato\"),\n", + " (\"a potato on top of a table\", \"there is a potato\"),\n", + " (\"a cat is sitting on a mat\", \"a mat is sitting on a cat\"),\n", + " (\"a dog chased the cat\", \"the cat chased the dog\"),\n", + " (\"a woman is holding a baby\", \"a baby is holding a woman\"),\n", + " (\"the boy kicked the ball\", \"the ball kicked the boy\"),\n", + " (\"a man is riding a horse\", \"a horse is riding a man\"),\n", + " (\"a girl is painting a picture\", \"a girl is creating art\"),\n", + " (\"two dogs are playing in a park\", \"animals are outdoors\"),\n", + " (\"a person is swimming in the ocean\", \"nobody is in the water\"),\n", + " ]\n", + "\n", + " with torch.no_grad():\n", + " for premise, hypothesis in test_pairs:\n", + " p_in = tokenizer([premise], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " h_in = tokenizer([hypothesis], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + " h_out = model(**h_in)\n", + "\n", + " p_feat = p_out.enriched if p_out.enriched is not None else p_out.last_hidden_state\n", + " h_feat = h_out.enriched if h_out.enriched is not None else h_out.last_hidden_state\n", + "\n", + " logits, _ = nli(p_feat, h_feat)\n", + " probs = F.softmax(logits, dim=-1)[0]\n", + " pred = label_names[probs.argmax()]\n", + "\n", + " cos = F.cosine_similarity(\n", + " p_out.last_hidden_state, h_out.last_hidden_state).item()\n", + "\n", + " print(f\"\\n P: {premise}\")\n", + " print(f\" H: {hypothesis}\")\n", + " print(f\" Pooled cos: {cos:.3f}\")\n", + " print(f\" NLI: {pred} [E={probs[0]:.3f} N={probs[1]:.3f} C={probs[2]:.3f}]\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Best val accuracy: {best_val_acc:.4f}\")\n", + " print(f\" Head params: {n_head_params:,}\")\n", + " print(f\" Paths: {len(nli.compositions)}\")\n", + " print(f\" Components: {nli.n_components} → d_path={nli.d_path}\")\n", + " print(f\" Bank present: {has_bank}\")\n", + " print(f\" Saved: nli_conv5d_best.pt\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "aaa1014322ee46b583d42f9c14926b69", + "5a2fc8ba9e1f40358f2ff2fd5a2417fb", + "699abe3252ac4a0eae550e70da47c4a8", + "d02dd92a6d854d0e8bcb67d313998f4a", + "a55ca004ee9641f8ade3bfa17b5aaa72", + "5203c99f5dce4b928172907f12fa3b66", + "f4bcfcda00c64a65b6b823d0f8395e1a", + "eb4d1532bbec441f877e19215516afa0", + "5fc8bc04e2214dcf81eb2524d533065e", + "2b7be4a063034e94aacabc7fad352d3d", + "5edb67365a5c4bef9c248a99febf42ca", + "db10dfddd4e942e1b64e0cda2d3aee64", + "c0396e0a06474078a37a6a84a19d8ce9", + "86ca7df7efda4534b017d17af36fb0e4", + "6a0cecf170354796ad0e64b6a1073c8b", + "085300d4121a445eaab1864861967ebf", + "aa4dd6dfb06242f19a25050e4eaacf4a", + "87cd43cc1bc140f99a0f826066a84a4e", + "70bfc00d11ef4b2aa607b449abf6b27f", + "6c6d9068a112407999016a2e90339756", + "538cdeaff560479ab2703f3810c4a69d", + "de747149f30a42c49fb8a11ed518364b" + ] + }, + "id": "s7gNF3zImn27", + "outputId": "4f286a88-b76f-454b-b36c-1a8801b6f01a" + }, + "execution_count": 15, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "NLI HEAD: Compositional Convolution (conv5d)\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "LOADING MODEL\n", + "=================================================================\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/112 [00:00 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " from transformers import AutoModel, AutoTokenizer\n", + " from datasets import load_dataset\n", + "\n", + " # ── Load model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING MODEL\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + " for p in model.parameters():\n", + " p.requires_grad = False\n", + "\n", + " has_bank = model.bank is not None\n", + " print(f\" Model: {sum(p.numel() for p in model.parameters()):,} params (frozen)\")\n", + " print(f\" Bank: {'present' if has_bank else 'absent'}\")\n", + "\n", + " # ── Load SNLI ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING SNLI\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ds = load_dataset(\"stanfordnlp/snli\")\n", + " train_ds = ds[\"train\"].filter(lambda x: x[\"label\"] >= 0)\n", + " val_ds = ds[\"validation\"].filter(lambda x: x[\"label\"] >= 0)\n", + " print(f\" Train: {len(train_ds):,} Val: {len(val_ds):,}\")\n", + "\n", + " MAX_TRAIN = 549000 # full SNLI\n", + " MAX_VAL = 9800\n", + "\n", + " # ── Pre-encode ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PRE-ENCODING\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " @torch.no_grad()\n", + " def encode_pairs(dataset, max_n, batch_size=1024):\n", + " dataset = dataset.select(range(min(max_n, len(dataset))))\n", + " all_p_enr, all_h_enr = [], []\n", + " all_labels = []\n", + "\n", + " for i in tqdm(range(0, len(dataset), batch_size), desc=\" Encoding\"):\n", + " j = min(i + batch_size, len(dataset))\n", + " batch = dataset[i:j]\n", + "\n", + " p_in = tokenizer(batch[\"premise\"], max_length=128,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + "\n", + " h_in = tokenizer(batch[\"hypothesis\"], max_length=128,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " h_out = model(**h_in)\n", + "\n", + " p_feat = p_out.enriched if p_out.enriched is not None else p_out.last_hidden_state\n", + " h_feat = h_out.enriched if h_out.enriched is not None else h_out.last_hidden_state\n", + "\n", + " all_p_enr.append(p_feat.cpu())\n", + " all_h_enr.append(h_feat.cpu())\n", + " all_labels.append(torch.tensor(batch[\"label\"]))\n", + "\n", + " return {\n", + " \"p\": torch.cat(all_p_enr),\n", + " \"h\": torch.cat(all_h_enr),\n", + " \"labels\": torch.cat(all_labels),\n", + " }\n", + "\n", + " train_data = encode_pairs(train_ds, MAX_TRAIN)\n", + " val_data = encode_pairs(val_ds, MAX_VAL)\n", + "\n", + " d_enriched = train_data[\"p\"].shape[1]\n", + " d_raw = 768\n", + " d_bank = d_enriched - d_raw\n", + " print(f\" Enriched: {d_enriched} (raw={d_raw} + bank={d_bank})\")\n", + " print(f\" Train: {train_data['labels'].shape[0]:,} Val: {val_data['labels'].shape[0]:,}\")\n", + "\n", + " for label, name in [(0, \"entailment\"), (1, \"neutral\"), (2, \"contradiction\")]:\n", + " n_l = (train_data[\"labels\"] == label).sum().item()\n", + " print(f\" {name}: {n_l:,} ({n_l/len(train_data['labels']):.1%})\")\n", + "\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # Move to GPU\n", + " train_p = train_data[\"p\"].to(DEVICE)\n", + " train_h = train_data[\"h\"].to(DEVICE)\n", + " train_labels = train_data[\"labels\"].to(DEVICE)\n", + " val_p = val_data[\"p\"].to(DEVICE)\n", + " val_h = val_data[\"h\"].to(DEVICE)\n", + " val_labels = val_data[\"labels\"].to(DEVICE)\n", + "\n", + " # ── Build CompConv NLI ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL CONV NLI HEAD\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli = CompConvNLI(\n", + " d_raw=d_raw, d_bank=max(d_bank, 1),\n", + " #d_path=512,\n", + " n_components=5, n_classes=3, dropout=0.3\n", + " ).to(DEVICE)\n", + " n_head_params = sum(p.numel() for p in nli.parameters())\n", + " print(f\" Head params: {n_head_params:,}\")\n", + "\n", + " # ── Training ──\n", + " EPOCHS = 20\n", + " BATCH = 128\n", + " LR = 1e-4\n", + " n_train = train_labels.shape[0]\n", + " n_batches = n_train // BATCH\n", + "\n", + " optimizer = torch.optim.AdamW(nli.parameters(), lr=LR, weight_decay=0.01)\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * EPOCHS, eta_min=1e-6)\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({EPOCHS} epochs, {n_batches} batches/epoch)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " best_val_acc = 0.0\n", + " for epoch in range(EPOCHS):\n", + " nli.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss, total_correct, n = 0, 0, 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + "\n", + " logits, _ = nli(train_p[idx], train_h[idx])\n", + " labels = train_labels[idx]\n", + " loss = F.cross_entropy(logits, labels)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(nli.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " total_correct += (logits.argmax(-1) == labels).sum().item()\n", + " total_loss += loss.item()\n", + " n += len(idx)\n", + "\n", + " elapsed = time.time() - t0\n", + " train_acc = total_correct / max(n, 1)\n", + " train_loss = total_loss / max(n // BATCH, 1)\n", + "\n", + " # Validation\n", + " nli.eval()\n", + " with torch.no_grad():\n", + " val_n = val_labels.shape[0]\n", + " val_correct = 0\n", + " val_loss_sum = 0\n", + " all_preds, all_labs = [], []\n", + " path_info = None\n", + "\n", + " for i in range(0, val_n, 512):\n", + " j = min(i + 512, val_n)\n", + " logits, info = nli(val_p[i:j], val_h[i:j])\n", + " labs = val_labels[i:j]\n", + " val_correct += (logits.argmax(-1) == labs).sum().item()\n", + " val_loss_sum += F.cross_entropy(logits, labs, reduction=\"sum\").item()\n", + " all_preds.append(logits.argmax(-1).cpu())\n", + " all_labs.append(labs.cpu())\n", + " if path_info is None:\n", + " path_info = info\n", + "\n", + " val_acc = val_correct / val_n\n", + " val_loss = val_loss_sum / val_n\n", + " preds = torch.cat(all_preds)\n", + " labs_all = torch.cat(all_labs)\n", + "\n", + " acc_ent = (preds[labs_all == 0] == 0).float().mean().item() if (labs_all == 0).sum() > 0 else 0\n", + " acc_neu = (preds[labs_all == 1] == 1).float().mean().item() if (labs_all == 1).sum() > 0 else 0\n", + " acc_con = (preds[labs_all == 2] == 2).float().mean().item() if (labs_all == 2).sum() > 0 else 0\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s\")\n", + " print(f\" Task: loss={train_loss:.4f} t_acc={train_acc:.4f} v_acc={val_acc:.4f} v_loss={val_loss:.4f}\")\n", + " print(f\" Per-class: ent={acc_ent:.3f} neu={acc_neu:.3f} con={acc_con:.3f}\")\n", + " if path_info:\n", + " top3 = path_info[\"top_paths\"][:3]\n", + " path_str = \" \".join(f\"{comp}={w:.3f}\" for comp, w in top3)\n", + " print(f\" Paths: {path_str} spread={path_info['weight_spread']:.4f}\")\n", + " print(f\" Protos: sim={path_info.get('proto_spread', 0):.4f} \"\n", + " f\"temp={path_info.get('temperature', 0):.2f}\")\n", + "\n", + " if val_acc > best_val_acc:\n", + " best_val_acc = val_acc\n", + " torch.save(nli.state_dict(), \"nli_conv5d_best.pt\")\n", + " print(f\" ★ New best: {val_acc:.4f}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # PATH ANALYSIS\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PATH WEIGHT ANALYSIS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli.load_state_dict(torch.load(\"nli_conv5d_best.pt\", weights_only=True))\n", + " nli.eval()\n", + "\n", + " weights = F.softmax(nli.path_weights, dim=0).cpu().tolist()\n", + " ranked = sorted(zip(nli.compositions, weights), key=lambda x: -x[1])\n", + " print(f\"\\n {'Path':<25} {'Weight':>8} {'Type':<15}\")\n", + " print(f\" {'-'*50}\")\n", + " for comp, w in ranked:\n", + " if len(comp) == 1:\n", + " ptype = \"holistic\"\n", + " elif all(c == 1 for c in comp):\n", + " ptype = \"independent\"\n", + " elif comp[0] >= 3:\n", + " ptype = \"geo-first\"\n", + " elif comp[0] == 1 and sum(comp[1:]) == 4:\n", + " ptype = \"geo→rest\"\n", + " elif comp[0] == 2:\n", + " ptype = \"geo+struct→...\"\n", + " else:\n", + " ptype = \"mixed\"\n", + " bar = \"█\" * int(w * 100)\n", + " print(f\" {str(comp):<25} {w:>8.4f} {ptype:<15} {bar}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # COMPOSITIONAL ORDER TEST\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL ORDER TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " label_names = [\"entailment\", \"neutral\", \"contradiction\"]\n", + "\n", + " test_pairs = [\n", + " (\"a potato on top of a table\", \"a table on top of a potato\"),\n", + " (\"a potato on top of a table\", \"there is a potato\"),\n", + " (\"a cat is sitting on a mat\", \"a mat is sitting on a cat\"),\n", + " (\"a dog chased the cat\", \"the cat chased the dog\"),\n", + " (\"a woman is holding a baby\", \"a baby is holding a woman\"),\n", + " (\"the boy kicked the ball\", \"the ball kicked the boy\"),\n", + " (\"a man is riding a horse\", \"a horse is riding a man\"),\n", + " (\"a girl is painting a picture\", \"a girl is creating art\"),\n", + " (\"two dogs are playing in a park\", \"animals are outdoors\"),\n", + " (\"a person is swimming in the ocean\", \"nobody is in the water\"),\n", + " ]\n", + "\n", + " with torch.no_grad():\n", + " for premise, hypothesis in test_pairs:\n", + " p_in = tokenizer([premise], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " h_in = tokenizer([hypothesis], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + " h_out = model(**h_in)\n", + "\n", + " p_feat = p_out.enriched if p_out.enriched is not None else p_out.last_hidden_state\n", + " h_feat = h_out.enriched if h_out.enriched is not None else h_out.last_hidden_state\n", + "\n", + " logits, _ = nli(p_feat, h_feat)\n", + " probs = F.softmax(logits, dim=-1)[0]\n", + " pred = label_names[probs.argmax()]\n", + "\n", + " cos = F.cosine_similarity(\n", + " p_out.last_hidden_state, h_out.last_hidden_state).item()\n", + "\n", + " print(f\"\\n P: {premise}\")\n", + " print(f\" H: {hypothesis}\")\n", + " print(f\" Pooled cos: {cos:.3f}\")\n", + " print(f\" NLI: {pred} [E={probs[0]:.3f} N={probs[1]:.3f} C={probs[2]:.3f}]\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Best val accuracy: {best_val_acc:.4f}\")\n", + " print(f\" Head params: {n_head_params:,}\")\n", + " print(f\" Paths: {len(nli.compositions)}\")\n", + " print(f\" Components: {nli.n_components} → d_path={nli.d_path}\")\n", + " print(f\" Bank present: {has_bank}\")\n", + " print(f\" Saved: nli_conv5d_best.pt\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "ef25e8cfba5845d1b4e302267a55af9e", + "a45a6e9a5f294549b6fc2526dabfc4f8", + "d0ddd09569cb402eb9293b6eb509911e", + "d418c749474048318eb434984e396faa", + "f3cc72fadce64c89808123ce201e9796", + "e4a2d1066d044474803ef1b433b4792f", + "67a56aac4c4d40ad8b109b707347ee45", + "eb0326c486d748248eaac16f09399b83", + "9323c9581fcc45eabf61356cdec62562", + "20139b20c7d44c41aa018faf16dc0eee", + "783f7928d25e4c40ac67d6a173090928", + "b23692bd70a0466a948ed71b0f41a845", + "b8edb67d8fe946b2b2f78a8b7b37bcfb", + "f6c31ab1a52d4cdfa3dc7528cc0fcf3c", + "73c89a1a107c43c58bd14f20defa058f", + "7a6a4867905f480ca2d43c02f2c198ea", + "22b1cb0cd8904ecc8f7dcfbca99d27a4", + "b124a1bc0ccc46e0bd10e03595415358", + "99567e7e1f994314bcf6a190e06a9817", + "6352aa71801b4e97889ddcab35a9e453", + "788e0cfd7db4483fb55591c77abb1720", + "5e814939a89d4afcb80484fac513b5e1" + ] + }, + "id": "9YnRBVweuZ-b", + "outputId": "bcc5086e-b986-4911-ad2c-46367ac48aa7" + }, + "execution_count": 2, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "NLI HEAD: Compositional Convolution (conv5d)\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "LOADING MODEL\n", + "=================================================================\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "A new version of the following files was downloaded from https://huggingface.co/AbstractPhil/geolip-captionbert-8192:\n", + "- modeling_caption_bert.py\n", + ". Make sure to double-check they do not contain any added malicious code. To avoid downloading new versions of the code file, you can pin a revision.\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/112 [00:00 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " from transformers import AutoModel, AutoTokenizer\n", + " from datasets import load_dataset\n", + "\n", + " # ── Load model ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING MODEL\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + " for p in model.parameters():\n", + " p.requires_grad = False\n", + "\n", + " has_bank = model.bank is not None\n", + " print(f\" Model: {sum(p.numel() for p in model.parameters()):,} params (frozen)\")\n", + " print(f\" Bank: {'present' if has_bank else 'absent'}\")\n", + "\n", + " # ── Load SNLI ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"LOADING SNLI\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " ds = load_dataset(\"stanfordnlp/snli\")\n", + " train_ds = ds[\"train\"].filter(lambda x: x[\"label\"] >= 0)\n", + " val_ds = ds[\"validation\"].filter(lambda x: x[\"label\"] >= 0)\n", + " print(f\" Train: {len(train_ds):,} Val: {len(val_ds):,}\")\n", + "\n", + " MAX_TRAIN = 549000 # full SNLI\n", + " MAX_VAL = 9800\n", + "\n", + " # ── Pre-encode ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PRE-ENCODING\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " @torch.no_grad()\n", + " def encode_pairs(dataset, max_n, batch_size=256):\n", + " dataset = dataset.select(range(min(max_n, len(dataset))))\n", + " all_p_enr, all_h_enr = [], []\n", + " all_labels = []\n", + "\n", + " for i in tqdm(range(0, len(dataset), batch_size), desc=\" Encoding\"):\n", + " j = min(i + batch_size, len(dataset))\n", + " batch = dataset[i:j]\n", + "\n", + " p_in = tokenizer(batch[\"premise\"], max_length=128,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + "\n", + " h_in = tokenizer(batch[\"hypothesis\"], max_length=128,\n", + " padding=\"max_length\", truncation=True,\n", + " return_tensors=\"pt\").to(DEVICE)\n", + " h_out = model(**h_in)\n", + "\n", + " p_feat = p_out.enriched if p_out.enriched is not None else p_out.last_hidden_state\n", + " h_feat = h_out.enriched if h_out.enriched is not None else h_out.last_hidden_state\n", + "\n", + " all_p_enr.append(p_feat.cpu())\n", + " all_h_enr.append(h_feat.cpu())\n", + " all_labels.append(torch.tensor(batch[\"label\"]))\n", + "\n", + " return {\n", + " \"p\": torch.cat(all_p_enr),\n", + " \"h\": torch.cat(all_h_enr),\n", + " \"labels\": torch.cat(all_labels),\n", + " }\n", + "\n", + " train_data = encode_pairs(train_ds, MAX_TRAIN)\n", + " val_data = encode_pairs(val_ds, MAX_VAL)\n", + "\n", + " d_enriched = train_data[\"p\"].shape[1]\n", + " d_raw = 768\n", + " d_bank = d_enriched - d_raw\n", + " print(f\" Enriched: {d_enriched} (raw={d_raw} + bank={d_bank})\")\n", + " print(f\" Train: {train_data['labels'].shape[0]:,} Val: {val_data['labels'].shape[0]:,}\")\n", + "\n", + " for label, name in [(0, \"entailment\"), (1, \"neutral\"), (2, \"contradiction\")]:\n", + " n_l = (train_data[\"labels\"] == label).sum().item()\n", + " print(f\" {name}: {n_l:,} ({n_l/len(train_data['labels']):.1%})\")\n", + "\n", + " del model; gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # Move to GPU\n", + " train_p = train_data[\"p\"].to(DEVICE)\n", + " train_h = train_data[\"h\"].to(DEVICE)\n", + " train_labels = train_data[\"labels\"].to(DEVICE)\n", + " val_p = val_data[\"p\"].to(DEVICE)\n", + " val_h = val_data[\"h\"].to(DEVICE)\n", + " val_labels = val_data[\"labels\"].to(DEVICE)\n", + "\n", + " # ── Build CompConv NLI ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL CONV NLI HEAD\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli = CompConvNLI(\n", + " d_raw=d_raw, d_bank=max(d_bank, 1),\n", + " d_path=256, n_components=5, n_classes=3, dropout=0.1\n", + " ).to(DEVICE)\n", + " n_head_params = sum(p.numel() for p in nli.parameters())\n", + " print(f\" Head params: {n_head_params:,}\")\n", + "\n", + " # ── Training ──\n", + " EPOCHS = 50\n", + " BATCH = 1024\n", + " LR = 1e-4\n", + " n_train = train_labels.shape[0]\n", + " n_batches = n_train // BATCH\n", + "\n", + " optimizer = torch.optim.AdamW(nli.parameters(), lr=LR, weight_decay=0.01)\n", + " scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=n_batches * EPOCHS, eta_min=1e-6)\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(f\"TRAINING ({EPOCHS} epochs, {n_batches} batches/epoch)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " best_val_acc = 0.0\n", + " for epoch in range(EPOCHS):\n", + " nli.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " total_loss, total_correct, n = 0, 0, 0\n", + " t0 = time.time()\n", + "\n", + " for i in range(0, n_train, BATCH):\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + "\n", + " logits, _ = nli(train_p[idx], train_h[idx])\n", + " labels = train_labels[idx]\n", + " loss = F.cross_entropy(logits, labels)\n", + "\n", + " loss.backward()\n", + " torch.nn.utils.clip_grad_norm_(nli.parameters(), 1.0)\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " total_correct += (logits.argmax(-1) == labels).sum().item()\n", + " total_loss += loss.item()\n", + " n += len(idx)\n", + "\n", + " elapsed = time.time() - t0\n", + " train_acc = total_correct / max(n, 1)\n", + " train_loss = total_loss / max(n // BATCH, 1)\n", + "\n", + " # Validation\n", + " nli.eval()\n", + " with torch.no_grad():\n", + " val_n = val_labels.shape[0]\n", + " val_correct = 0\n", + " val_loss_sum = 0\n", + " all_preds, all_labs = [], []\n", + " path_info = None\n", + "\n", + " for i in range(0, val_n, 512):\n", + " j = min(i + 512, val_n)\n", + " logits, info = nli(val_p[i:j], val_h[i:j])\n", + " labs = val_labels[i:j]\n", + " val_correct += (logits.argmax(-1) == labs).sum().item()\n", + " val_loss_sum += F.cross_entropy(logits, labs, reduction=\"sum\").item()\n", + " all_preds.append(logits.argmax(-1).cpu())\n", + " all_labs.append(labs.cpu())\n", + " if path_info is None:\n", + " path_info = info\n", + "\n", + " val_acc = val_correct / val_n\n", + " val_loss = val_loss_sum / val_n\n", + " preds = torch.cat(all_preds)\n", + " labs_all = torch.cat(all_labs)\n", + "\n", + " acc_ent = (preds[labs_all == 0] == 0).float().mean().item() if (labs_all == 0).sum() > 0 else 0\n", + " acc_neu = (preds[labs_all == 1] == 1).float().mean().item() if (labs_all == 1).sum() > 0 else 0\n", + " acc_con = (preds[labs_all == 2] == 2).float().mean().item() if (labs_all == 2).sum() > 0 else 0\n", + "\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s\")\n", + " print(f\" Task: loss={train_loss:.4f} t_acc={train_acc:.4f} v_acc={val_acc:.4f} v_loss={val_loss:.4f}\")\n", + " print(f\" Per-class: ent={acc_ent:.3f} neu={acc_neu:.3f} con={acc_con:.3f}\")\n", + " if path_info:\n", + " top3 = path_info[\"top_paths\"][:3]\n", + " path_str = \" \".join(f\"{comp}={w:.3f}\" for comp, w in top3)\n", + " print(f\" Paths: {path_str} spread={path_info['weight_spread']:.4f}\")\n", + " print(f\" Protos: sim={path_info.get('proto_spread', 0):.4f} \"\n", + " f\"temp={path_info.get('temperature', 0):.2f}\")\n", + "\n", + " if val_acc > best_val_acc:\n", + " best_val_acc = val_acc\n", + " torch.save(nli.state_dict(), \"nli_conv5d_best.pt\")\n", + " print(f\" ★ New best: {val_acc:.4f}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # PATH ANALYSIS\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PATH WEIGHT ANALYSIS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " nli.load_state_dict(torch.load(\"nli_conv5d_best.pt\", weights_only=True))\n", + " nli.eval()\n", + "\n", + " weights = F.softmax(nli.path_weights, dim=0).cpu().tolist()\n", + " ranked = sorted(zip(nli.compositions, weights), key=lambda x: -x[1])\n", + " print(f\"\\n {'Path':<25} {'Weight':>8} {'Type':<15}\")\n", + " print(f\" {'-'*50}\")\n", + " for comp, w in ranked:\n", + " if len(comp) == 1:\n", + " ptype = \"holistic\"\n", + " elif all(c == 1 for c in comp):\n", + " ptype = \"independent\"\n", + " elif comp[0] >= 3:\n", + " ptype = \"geo-first\"\n", + " elif comp[0] == 1 and sum(comp[1:]) == 4:\n", + " ptype = \"geo→rest\"\n", + " elif comp[0] == 2:\n", + " ptype = \"geo+struct→...\"\n", + " else:\n", + " ptype = \"mixed\"\n", + " bar = \"█\" * int(w * 100)\n", + " print(f\" {str(comp):<25} {w:>8.4f} {ptype:<15} {bar}\")\n", + "\n", + " # ══════════════════════════════════════════════════════════════\n", + " # COMPOSITIONAL ORDER TEST\n", + " # ══════════════════════════════════════════════════════════════\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"COMPOSITIONAL ORDER TEST\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE).eval()\n", + " label_names = [\"entailment\", \"neutral\", \"contradiction\"]\n", + "\n", + " test_pairs = [\n", + " (\"a potato on top of a table\", \"a table on top of a potato\"),\n", + " (\"a potato on top of a table\", \"there is a potato\"),\n", + " (\"a cat is sitting on a mat\", \"a mat is sitting on a cat\"),\n", + " (\"a dog chased the cat\", \"the cat chased the dog\"),\n", + " (\"a woman is holding a baby\", \"a baby is holding a woman\"),\n", + " (\"the boy kicked the ball\", \"the ball kicked the boy\"),\n", + " (\"a man is riding a horse\", \"a horse is riding a man\"),\n", + " (\"a girl is painting a picture\", \"a girl is creating art\"),\n", + " (\"two dogs are playing in a park\", \"animals are outdoors\"),\n", + " (\"a person is swimming in the ocean\", \"nobody is in the water\"),\n", + " ]\n", + "\n", + " with torch.no_grad():\n", + " for premise, hypothesis in test_pairs:\n", + " p_in = tokenizer([premise], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " h_in = tokenizer([hypothesis], max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " p_out = model(**p_in)\n", + " h_out = model(**h_in)\n", + "\n", + " p_feat = p_out.enriched if p_out.enriched is not None else p_out.last_hidden_state\n", + " h_feat = h_out.enriched if h_out.enriched is not None else h_out.last_hidden_state\n", + "\n", + " logits, _ = nli(p_feat, h_feat)\n", + " probs = F.softmax(logits, dim=-1)[0]\n", + " pred = label_names[probs.argmax()]\n", + "\n", + " cos = F.cosine_similarity(\n", + " p_out.last_hidden_state, h_out.last_hidden_state).item()\n", + "\n", + " print(f\"\\n P: {premise}\")\n", + " print(f\" H: {hypothesis}\")\n", + " print(f\" Pooled cos: {cos:.3f}\")\n", + " print(f\" NLI: {pred} [E={probs[0]:.3f} N={probs[1]:.3f} C={probs[2]:.3f}]\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Best val accuracy: {best_val_acc:.4f}\")\n", + " print(f\" Head params: {n_head_params:,}\")\n", + " print(f\" Paths: {len(nli.compositions)}\")\n", + " print(f\" Components: {nli.n_components} → d_path={nli.d_path}\")\n", + " print(f\" Bank present: {has_bank}\")\n", + " print(f\" Saved: nli_conv5d_best.pt\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "27627370b8284c5a862ffd5a3f953352", + "338ff68f8d7445e7bfdb3cbbf8716d86", + "449f0df7844148c59c86f1f6eccf2197", + "a293658c27454a47aadd082610e78d06", + "1f05373fe992499995c535ec94ab4404", + "606a421f64114ed0aed8fc0437f0c171", + "787e8cc4638e42318b2b93976b04893d", + "5a76e50e8636432b85c6031fcb303f03", + "c537d231252e4684a0727ec9171e696c", + "da8eea806fcb4c9cbd0dbc3d1c6de5ad", + "18edc062bd4d4c54a496b8c92e2f9d99", + "eef3768972ff4b839344f03c681e057d", + "a12a526b205c422dbc9d51765b12c2da", + "df4b6755e47845538d30410ee1ffaef3", + "88c92b8203c8464897ffec2683f1724b", + "db0cfa00a74749eab32a13a3e81e9426", + "4831efd2dcb14557af97e6f627a932ec", + "08b0d9679dc24fa2a86a874c60b67d52", + "99a3d94fb5dc457da2fa3968e51f2345", + "15a95ffece5e419ab00699f951a71c03", + "741ee9ffa3084edea972f8f09a877062", + "d1a3d362dbf9426b9b2b8bb95b0abf52" + ] + }, + "id": "EILZHZu4OE7v", + "outputId": "cce90cb9-86d1-4614-b1ca-57e7fd67f25f" + }, + "execution_count": 5, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "NLI HEAD: Compositional Convolution (conv5d)\n", + "=================================================================\n", + " Device: cuda\n", + "\n", + "=================================================================\n", + "LOADING MODEL\n", + "=================================================================\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/112 [00:00 0: vols.append(v)\n", + " if len(vols) < 10: return 0.0\n", + " a = np.array(vols)\n", + " return float(a.std() / (a.mean() + 1e-8))\n", + "\n", + "def infonce(a, b, temperature=0.07):\n", + " a = F.normalize(a, dim=-1)\n", + " b = F.normalize(b, dim=-1)\n", + " logits = (a @ b.T) / temperature\n", + " labels = torch.arange(logits.shape[0], device=logits.device)\n", + " loss = (F.cross_entropy(logits, labels) + F.cross_entropy(logits.T, labels)) / 2\n", + " with torch.no_grad():\n", + " acc = (logits.argmax(-1) == labels).float().mean().item()\n", + " return loss, acc\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# BANK LOSS (differentiable, computed externally)\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def compute_bank_loss(bank, embedding):\n", + " \"\"\"\n", + " Full refined bank loss — matches production train_alignment_bank.py exactly.\n", + " All targets from bank's calibrated buffers. No hardcoded constants.\n", + " Differentiable throughout — no .item() calls on loss components.\n", + " \"\"\"\n", + " B = embedding.shape[0]\n", + " emb = embedding.float()\n", + "\n", + " # ── Per-expert full whitened Procrustes ──\n", + " expert_cos_list = []\n", + " expert_projected = []\n", + " for i in range(bank.n_experts):\n", + " R = bank.expert_rotations[i]\n", + " W = bank.expert_whiteners[i]\n", + " mu = bank.expert_means[i]\n", + " centered = emb - mu\n", + " whitened = centered @ W\n", + " whitened_n = F.normalize(whitened, dim=-1)\n", + " in_expert = whitened_n @ R.T\n", + " back = in_expert @ R\n", + " cos = F.cosine_similarity(whitened_n, back, dim=-1)\n", + " expert_cos_list.append(cos)\n", + " expert_projected.append(in_expert)\n", + "\n", + " expert_cos = torch.stack(expert_cos_list, dim=-1)\n", + "\n", + " # ── 1. Expert agreement ──\n", + " expert_mean = expert_cos.mean(dim=-1, keepdim=True)\n", + " l_agreement = (expert_cos - expert_mean).pow(2).mean()\n", + "\n", + " # ── 2. Rotation orthogonality ──\n", + " l_ortho = 0.0\n", + " for i in range(bank.n_experts):\n", + " R = bank.expert_rotations[i]\n", + " l_ortho += (R @ R.T - torch.eye(bank.d_embed, device=R.device)).pow(2).mean()\n", + " l_ortho = l_ortho / bank.n_experts\n", + "\n", + " # ── 3. Anchor spread ──\n", + " anchors_n = F.normalize(bank.anchors, dim=-1)\n", + " anchor_sim = anchors_n @ anchors_n.T\n", + " anchor_sim.fill_diagonal_(0)\n", + " l_spread = anchor_sim.pow(2).mean()\n", + "\n", + " # ── 4. Anchor entropy (sharpness) ──\n", + " anchor_cos = emb @ anchors_n.T\n", + " anchor_probs = F.softmax(anchor_cos * 10, dim=-1)\n", + " l_entropy = -(anchor_probs * (anchor_probs + 1e-12).log()).sum(-1).mean()\n", + "\n", + " # ── 5. Cross-expert differentiation ──\n", + " cross_cos = []\n", + " for i in range(bank.n_experts):\n", + " for j in range(i + 1, bank.n_experts):\n", + " cc = F.cosine_similarity(expert_projected[i], expert_projected[j], dim=-1)\n", + " cross_cos.append(cc)\n", + "\n", + " if cross_cos:\n", + " cross_features = torch.stack(cross_cos, dim=-1)\n", + " l_cross_var = cross_features.var(dim=0).mean()\n", + "\n", + " # ── 6. Disagreement preservation (cross_cos + ratio) ──\n", + " batch_cross_mean = cross_features.mean()\n", + " batch_cross_std = cross_features.std()\n", + " per_sample_agreement = expert_cos.mean(dim=-1)\n", + " per_sample_disagreement = expert_cos.std(dim=-1)\n", + " batch_disagree_ratio = (per_sample_disagreement / (per_sample_agreement + 1e-8)).mean()\n", + " l_disagree = (\n", + " (batch_cross_mean - bank.target_cross_cos_mean).pow(2) +\n", + " (batch_cross_std - bank.target_cross_cos_std).pow(2) +\n", + " (batch_disagree_ratio - bank.target_disagreement_ratio).pow(2)\n", + " )\n", + " else:\n", + " l_cross_var = torch.tensor(0.0, device=emb.device)\n", + " l_disagree = torch.tensor(0.0, device=emb.device)\n", + "\n", + " # ── 7. Embedding CV → should track bank.target_cv ──\n", + " l_emb_cv = torch.tensor(0.0, device=emb.device)\n", + " if B >= 10:\n", + " emb_n = F.normalize(emb, dim=-1)\n", + " vols = []\n", + " for _ in range(16):\n", + " idx = torch.randperm(B, device=emb.device)[:5]\n", + " pts = emb_n[idx].unsqueeze(0)\n", + " diff = pts.unsqueeze(-2) - pts.unsqueeze(-3)\n", + " d2 = (diff * diff).sum(-1)\n", + " Bv, V, _ = d2.shape\n", + " cm = torch.zeros(Bv, V+1, V+1, device=d2.device, dtype=torch.float32)\n", + " cm[:, 0, 1:] = 1; cm[:, 1:, 0] = 1; cm[:, 1:, 1:] = d2\n", + " s = (-1.0)**V; f = math.factorial(V-1)\n", + " v2 = s / ((2.0**(V-1)) * f*f) * torch.linalg.det(cm)\n", + " vols.append(torch.sqrt(F.relu(v2[0]) + 1e-12))\n", + " stacked = torch.stack(vols)\n", + " emb_cv = stacked.std() / (stacked.mean() + 1e-8)\n", + " l_emb_cv = (emb_cv - bank.target_cv).abs()\n", + "\n", + " # ── Combined (same weights as production bank trainer) ──\n", + " total = (1.0 * l_agreement +\n", + " 1.0 * l_ortho +\n", + " 0.5 * l_spread +\n", + " 0.1 * l_entropy +\n", + " 0.3 * l_cross_var +\n", + " 0.3 * l_emb_cv +\n", + " 0.5 * l_disagree)\n", + "\n", + " diagnostics = {\n", + " \"agreement\": l_agreement.item(),\n", + " \"ortho\": l_ortho.item() if torch.is_tensor(l_ortho) else l_ortho,\n", + " \"spread\": l_spread.item(),\n", + " \"entropy\": l_entropy.item(),\n", + " \"cross_var\": l_cross_var.item(),\n", + " \"disagree\": l_disagree.item(),\n", + " \"emb_cv\": emb_cv.item() if B >= 10 else 0.0,\n", + " \"expert_cos_mean\": expert_cos.mean().item(),\n", + " \"expert_cos_std\": expert_cos.std().item(),\n", + " }\n", + "\n", + " return total, diagnostics\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# ALIGNMENT UTILITIES\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def symmetric_inv_sqrt(cov, eps=1e-6):\n", + " evals, evecs = torch.linalg.eigh(cov)\n", + " evals = torch.clamp(evals, min=eps)\n", + " return evecs @ torch.diag(evals.rsqrt()) @ evecs.T\n", + "\n", + "def procrustes_align(source, target, n_align=10000):\n", + " N = min(n_align, source.shape[0], target.shape[0])\n", + " S = source[:N].float(); T = target[:N].float()\n", + " s_mean = S.mean(0, keepdim=True); t_mean = T.mean(0, keepdim=True)\n", + " Sc = S - s_mean; Tc = T - t_mean; N_s = Sc.shape[0]\n", + " cos_before = F.cosine_similarity(Sc, Tc, dim=-1).mean().item()\n", + " s_cov = (Sc.T @ Sc) / max(N_s - 1, 1)\n", + " t_cov = (Tc.T @ Tc) / max(N_s - 1, 1)\n", + " s_whiten = symmetric_inv_sqrt(s_cov)\n", + " t_whiten = symmetric_inv_sqrt(t_cov)\n", + " Sc_w = F.normalize(Sc @ s_whiten, dim=-1)\n", + " Tc_w = F.normalize(Tc @ t_whiten, dim=-1)\n", + " U, _, Vt = torch.linalg.svd(Tc_w.T @ Sc_w, full_matrices=False)\n", + " R = U @ Vt\n", + " cos_after = F.cosine_similarity(Sc_w @ R.T, Tc_w, dim=-1).mean().item()\n", + " return {\"rotation\": R, \"source_mean\": s_mean.squeeze(0),\n", + " \"source_whitener\": s_whiten,\n", + " \"target_unwhitener\": torch.linalg.pinv(t_whiten),\n", + " \"cos_before\": cos_before, \"cos_after\": cos_after}\n", + "\n", + "def apply_align(emb, a):\n", + " x = emb.float() - a[\"source_mean\"]\n", + " x = x @ a[\"source_whitener\"]; x = x @ a[\"rotation\"].T\n", + " x = x @ a[\"target_unwhitener\"]; return x\n", + "\n", + "\n", + "# ══════════════════════════════════════════════════════════════════\n", + "# MAIN\n", + "# ══════════════════════════════════════════════════════════════════\n", + "\n", + "def run():\n", + " torch.manual_seed(42)\n", + " np.random.seed(42)\n", + " names = [\"bert\", \"modern\", \"roberta\", \"albert\", \"distil\"]\n", + "\n", + " # ── Phase 0: Load cached embeddings ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 0: LOAD CACHED EMBEDDINGS\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " embeds = {}\n", + " for name in names:\n", + " p = os.path.join(CACHE_DIR, f\"{name}.pt\")\n", + " embeds[name] = torch.load(p, weights_only=True)\n", + " print(f\" {name}: {embeds[name].shape}\")\n", + "\n", + " caps_path = os.path.join(CACHE_DIR, \"captions.json\")\n", + " with open(caps_path) as f:\n", + " captions = json.load(f)\n", + " N = min(len(captions), min(e.shape[0] for e in embeds.values()))\n", + " print(f\" Captions: {len(captions):,}, using {N:,}\")\n", + "\n", + " # ── Phase 1: GPA Alignment ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 1: GPA ALIGNMENT\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " current = {name: embeds[name][:N].float() for name in names}\n", + " for gpa_iter in range(15):\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + " total_delta = 0.0\n", + " new_current = {}\n", + " for name in names:\n", + " info = procrustes_align(current[name], mean_shape)\n", + " new_current[name] = apply_align(current[name], info)\n", + " total_delta += (new_current[name] - current[name]).pow(2).mean().item()\n", + " current = new_current\n", + " if gpa_iter == 0 or (gpa_iter + 1) % 5 == 0:\n", + " print(f\" GPA iter {gpa_iter+1}: delta={total_delta:.8f}\")\n", + " if total_delta < 1e-8:\n", + " print(f\" Converged at iteration {gpa_iter+1}\")\n", + " break\n", + "\n", + " mean_shape = sum(current[n] for n in names) / len(names)\n", + " aligned = {}\n", + " for name in names:\n", + " info = procrustes_align(embeds[name][:N], mean_shape)\n", + " aligned[name] = apply_align(embeds[name][:N], info)\n", + "\n", + " consensus = F.normalize(sum(aligned[n] for n in names) / len(names), dim=-1)\n", + " for name in names:\n", + " c = F.cosine_similarity(consensus[:5000], aligned[name][:5000], dim=-1).mean().item()\n", + " print(f\" cos(consensus, {name}): {c:.4f}\")\n", + "\n", + " # Consensus CV\n", + " consensus_cv = cv_metric(consensus[:5000].to(DEVICE))\n", + " print(f\" Consensus CV: {consensus_cv:.4f}\")\n", + "\n", + " del embeds, aligned, current, mean_shape\n", + " gc.collect()\n", + "\n", + " # ── Phase 2: Load model (unfrozen) ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 2: LOAD MODEL (unfrozen)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " from transformers import AutoModel, AutoTokenizer\n", + "\n", + " model = AutoModel.from_pretrained(REPO_ID, trust_remote_code=True).to(DEVICE)\n", + " tokenizer = AutoTokenizer.from_pretrained(REPO_ID, trust_remote_code=True)\n", + "\n", + " has_bank = model.bank is not None\n", + " n_encoder = sum(p.numel() for n, p in model.named_parameters() if not n.startswith(\"bank.\"))\n", + " n_bank = sum(p.numel() for n, p in model.named_parameters() if n.startswith(\"bank.\"))\n", + " print(f\" Encoder: {n_encoder:,} params\")\n", + " print(f\" Bank: {n_bank:,} params ({'present' if has_bank else 'absent'})\")\n", + " print(f\" Total: {n_encoder + n_bank:,} params (ALL unfrozen)\")\n", + "\n", + " # ── Phase 3: Prepare data ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 3: TOKENIZE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " captions = captions[:N]\n", + " print(f\" Tokenizing {N:,} captions...\")\n", + " tokens = tokenizer(captions, max_length=512, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\")\n", + " input_ids = tokens[\"input_ids\"]\n", + " attention_mask = tokens[\"attention_mask\"]\n", + "\n", + " n_train = N - N_VAL\n", + " train_ids = input_ids[:n_train].to(DEVICE)\n", + " train_mask = attention_mask[:n_train].to(DEVICE)\n", + " train_targets = consensus[:n_train].to(DEVICE)\n", + " val_ids = input_ids[n_train:n_train + N_VAL].to(DEVICE)\n", + " val_mask = attention_mask[n_train:n_train + N_VAL].to(DEVICE)\n", + " val_targets = consensus[n_train:n_train + N_VAL].to(DEVICE)\n", + " print(f\" Train: {n_train:,} Val: {N_VAL:,}\")\n", + "\n", + " del tokens, input_ids, attention_mask\n", + " gc.collect(); torch.cuda.empty_cache()\n", + "\n", + " # ── Phase 4: Training ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 4: CO-TRAIN (encoder + bank)\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " # Separate param groups: encoder gets lower LR\n", + " encoder_params = [p for n, p in model.named_parameters() if not n.startswith(\"bank.\")]\n", + " bank_params = [p for n, p in model.named_parameters() if n.startswith(\"bank.\")]\n", + "\n", + " optimizer = torch.optim.AdamW([\n", + " {\"params\": encoder_params, \"lr\": LR_ENCODER},\n", + " {\"params\": bank_params, \"lr\": LR_BANK},\n", + " ], weight_decay=WEIGHT_DECAY)\n", + "\n", + " total_steps = (n_train // BATCH) * EPOCHS\n", + " scheduler = torch.optim.lr_scheduler.SequentialLR(\n", + " optimizer,\n", + " [torch.optim.lr_scheduler.LinearLR(optimizer, start_factor=0.01,\n", + " total_iters=WARMUP_STEPS),\n", + " torch.optim.lr_scheduler.CosineAnnealingLR(\n", + " optimizer, T_max=max(total_steps - WARMUP_STEPS, 1), eta_min=1e-6)],\n", + " milestones=[WARMUP_STEPS])\n", + "\n", + " best_val_cos = 0.0\n", + " global_step = 0\n", + " log_dir = f\"runs/cotrain_{time.strftime('%Y%m%d_%H%M%S')}\"\n", + " writer = SummaryWriter(log_dir)\n", + " print(f\" Tensorboard: {log_dir}\")\n", + "\n", + " # Log hyperparameters\n", + " writer.add_text(\"config/model\", f\"encoder={n_encoder:,} bank={n_bank:,} total={n_encoder+n_bank:,}\")\n", + " writer.add_text(\"config/training\", f\"epochs={EPOCHS} batch={BATCH} lr_enc={LR_ENCODER} lr_bank={LR_BANK}\")\n", + " writer.add_text(\"config/losses\", f\"nce={W_NCE} mse={W_MSE} cv={W_CV} bank={W_BANK}\")\n", + " writer.add_text(\"config/consensus\", f\"cv={consensus_cv:.4f}\")\n", + "\n", + " # Log initial bank state\n", + " if has_bank:\n", + " writer.add_scalar(\"bank_targets/cv\", model.bank.target_cv.item(), 0)\n", + " writer.add_scalar(\"bank_targets/cross_cos_mean\", model.bank.target_cross_cos_mean.item(), 0)\n", + " writer.add_scalar(\"bank_targets/cross_cos_std\", model.bank.target_cross_cos_std.item(), 0)\n", + " writer.add_scalar(\"bank_targets/disagreement_ratio\", model.bank.target_disagreement_ratio.item(), 0)\n", + "\n", + " for epoch in range(EPOCHS):\n", + " model.train()\n", + " perm = torch.randperm(n_train, device=DEVICE)\n", + " n = 0\n", + " t0 = time.time()\n", + "\n", + " pbar = tqdm(range(0, n_train, BATCH),\n", + " desc=f\"E{epoch+1:2d}/{EPOCHS}\", unit=\"batch\")\n", + " for i in pbar:\n", + " idx = perm[i:i+BATCH]\n", + " if len(idx) < 8: continue\n", + "\n", + " out = model(train_ids[idx], train_mask[idx])\n", + " emb = out.last_hidden_state\n", + " tgt = train_targets[idx]\n", + "\n", + " # Student losses\n", + " l_nce, acc = infonce(emb, tgt)\n", + " l_mse = F.mse_loss(emb, tgt)\n", + " l_cv = cv_loss(emb, target=consensus_cv)\n", + "\n", + " # Bank losses\n", + " if has_bank:\n", + " l_bank, bank_diag = compute_bank_loss(model.bank, emb)\n", + " else:\n", + " l_bank = torch.tensor(0.0, device=DEVICE)\n", + " bank_diag = {}\n", + "\n", + " loss = (W_NCE * l_nce + W_MSE * l_mse +\n", + " W_CV * l_cv + W_BANK * l_bank)\n", + "\n", + " loss.backward()\n", + " # Track gradient norms before clipping\n", + " with torch.no_grad():\n", + " enc_grad_norm = torch.nn.utils.clip_grad_norm_(\n", + " [p for n, p in model.named_parameters() if not n.startswith(\"bank.\")],\n", + " GRAD_CLIP)\n", + " bank_grad_norm = torch.nn.utils.clip_grad_norm_(\n", + " [p for n, p in model.named_parameters() if n.startswith(\"bank.\")],\n", + " GRAD_CLIP) if has_bank else 0.0\n", + "\n", + " optimizer.step(); optimizer.zero_grad(set_to_none=True)\n", + " scheduler.step()\n", + "\n", + " with torch.no_grad():\n", + " cos = F.cosine_similarity(emb, tgt, dim=-1).mean().item()\n", + " cos_std = F.cosine_similarity(emb, tgt, dim=-1).std().item()\n", + "\n", + " global_step += 1\n", + " n += 1\n", + "\n", + " # ── Step-level tensorboard (every 50 steps) ──\n", + " if global_step % 50 == 0:\n", + " # Losses\n", + " writer.add_scalar(\"train_loss/total\", loss.item(), global_step)\n", + " writer.add_scalar(\"train_loss/nce\", l_nce.item(), global_step)\n", + " writer.add_scalar(\"train_loss/mse\", l_mse.item(), global_step)\n", + " writer.add_scalar(\"train_loss/cv\", l_cv.item(), global_step)\n", + " writer.add_scalar(\"train_loss/bank\", l_bank.item() if torch.is_tensor(l_bank) else l_bank, global_step)\n", + "\n", + " # Student metrics\n", + " writer.add_scalar(\"train_student/cosine_mean\", cos, global_step)\n", + " writer.add_scalar(\"train_student/cosine_std\", cos_std, global_step)\n", + " writer.add_scalar(\"train_student/retrieval_acc\", acc, global_step)\n", + "\n", + " # Gradient norms\n", + " writer.add_scalar(\"gradients/encoder_norm\", enc_grad_norm, global_step)\n", + " if has_bank:\n", + " writer.add_scalar(\"gradients/bank_norm\", bank_grad_norm, global_step)\n", + "\n", + " # Learning rates\n", + " writer.add_scalar(\"lr/encoder\", optimizer.param_groups[0][\"lr\"], global_step)\n", + " if len(optimizer.param_groups) > 1:\n", + " writer.add_scalar(\"lr/bank\", optimizer.param_groups[1][\"lr\"], global_step)\n", + "\n", + " # Bank diagnostics\n", + " if bank_diag:\n", + " writer.add_scalar(\"train_bank/expert_agreement\", bank_diag[\"agreement\"], global_step)\n", + " writer.add_scalar(\"train_bank/rotation_ortho\", bank_diag[\"ortho\"], global_step)\n", + " writer.add_scalar(\"train_bank/anchor_spread\", bank_diag[\"spread\"], global_step)\n", + " writer.add_scalar(\"train_bank/anchor_entropy\", bank_diag[\"entropy\"], global_step)\n", + " writer.add_scalar(\"train_bank/cross_expert_var\", bank_diag[\"cross_var\"], global_step)\n", + " writer.add_scalar(\"train_bank/disagreement_preserve\", bank_diag[\"disagree\"], global_step)\n", + " writer.add_scalar(\"train_bank/emb_cv\", bank_diag[\"emb_cv\"], global_step)\n", + " writer.add_scalar(\"train_bank/expert_cos_mean\", bank_diag[\"expert_cos_mean\"], global_step)\n", + " writer.add_scalar(\"train_bank/expert_cos_std\", bank_diag[\"expert_cos_std\"], global_step)\n", + "\n", + " # ── Detailed step logging (every 500 steps) ──\n", + " if global_step % 500 == 0 and has_bank:\n", + " with torch.no_grad():\n", + " # Per-expert round-trip cosines\n", + " for exp_i in range(model.bank.n_experts):\n", + " R = model.bank.expert_rotations[exp_i]\n", + " W = model.bank.expert_whiteners[exp_i]\n", + " mu = model.bank.expert_means[exp_i]\n", + " centered = emb.float() - mu\n", + " whitened = centered @ W\n", + " whitened_n = F.normalize(whitened, dim=-1)\n", + " in_expert = whitened_n @ R.T\n", + " back = in_expert @ R\n", + " exp_cos = F.cosine_similarity(whitened_n, back, dim=-1).mean().item()\n", + " writer.add_scalar(f\"train_per_expert/cos_expert_{exp_i}\", exp_cos, global_step)\n", + "\n", + " # Rotation condition number\n", + " sv = torch.linalg.svdvals(R)\n", + " cond = (sv.max() / (sv.min() + 1e-8)).item()\n", + " writer.add_scalar(f\"train_per_expert/rotation_cond_{exp_i}\", cond, global_step)\n", + "\n", + " # Cross-expert cosine matrix (10 pairs)\n", + " expert_projected = []\n", + " for exp_i in range(model.bank.n_experts):\n", + " R = model.bank.expert_rotations[exp_i]\n", + " W = model.bank.expert_whiteners[exp_i]\n", + " mu = model.bank.expert_means[exp_i]\n", + " wn = F.normalize((emb.float() - mu) @ W, dim=-1)\n", + " expert_projected.append(wn @ R.T)\n", + "\n", + " pair_idx = 0\n", + " expert_names = [\"bert\", \"modern\", \"roberta\", \"albert\", \"distil\"]\n", + " for ei in range(model.bank.n_experts):\n", + " for ej in range(ei + 1, model.bank.n_experts):\n", + " cc = F.cosine_similarity(\n", + " expert_projected[ei], expert_projected[ej], dim=-1).mean().item()\n", + " writer.add_scalar(\n", + " f\"train_cross_expert/{expert_names[ei]}_{expert_names[ej]}\", cc, global_step)\n", + " pair_idx += 1\n", + "\n", + " # Anchor diagnostics\n", + " anchors_n = F.normalize(model.bank.anchors, dim=-1)\n", + " anchor_cos = emb.float() @ anchors_n.T\n", + " writer.add_scalar(\"train_anchors/max_cos\", anchor_cos.max(dim=-1).values.mean().item(), global_step)\n", + " writer.add_scalar(\"train_anchors/mean_cos\", anchor_cos.mean().item(), global_step)\n", + " writer.add_scalar(\"train_anchors/top3_mean\",\n", + " anchor_cos.topk(3, dim=-1).values.mean().item(), global_step)\n", + "\n", + " # Embedding space diagnostics\n", + " pairwise = emb @ emb.T\n", + " mask = ~torch.eye(emb.shape[0], dtype=torch.bool, device=DEVICE)\n", + " writer.add_scalar(\"train_embedding/pairwise_cos_mean\", pairwise[mask].mean().item(), global_step)\n", + " writer.add_scalar(\"train_embedding/pairwise_cos_std\", pairwise[mask].std().item(), global_step)\n", + "\n", + " # Spectral (on batch)\n", + " centered_emb = emb.float() - emb.float().mean(0, keepdim=True)\n", + " sv_emb = torch.linalg.svdvals(centered_emb)\n", + " eff_dim = float((sv_emb.sum() ** 2) / (sv_emb.pow(2).sum() + 1e-12))\n", + " writer.add_scalar(\"train_embedding/effective_dim\", eff_dim, global_step)\n", + " writer.add_scalar(\"train_embedding/spectral_top1\",\n", + " (sv_emb[0] / (sv_emb.sum() + 1e-8)).item(), global_step)\n", + "\n", + " d = max(n, 1)\n", + " pbar.set_postfix(loss=f\"{loss.item():.4f}\",\n", + " cos=f\"{cos:.3f}\",\n", + " bank=f\"{l_bank.item() if torch.is_tensor(l_bank) else 0:.4f}\")\n", + " pbar.close()\n", + "\n", + " elapsed = time.time() - t0\n", + "\n", + " # ── Epoch-level validation ──\n", + " model.eval()\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, N_VAL, 512):\n", + " vj = min(vi + 512, N_VAL)\n", + " vo = model(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(vo.last_hidden_state)\n", + " val_emb = torch.cat(val_embs)\n", + " _, val_acc = infonce(val_emb[:2000], val_targets[:2000])\n", + " val_cos = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " val_cos_std = F.cosine_similarity(val_emb, val_targets, dim=-1).std().item()\n", + " val_cv = cv_metric(val_emb[:2000])\n", + "\n", + " # Val spectral\n", + " val_centered = val_emb[:2000].float() - val_emb[:2000].float().mean(0, keepdim=True)\n", + " val_sv = torch.linalg.svdvals(val_centered)\n", + " val_eff_dim = float((val_sv.sum() ** 2) / (val_sv.pow(2).sum() + 1e-12))\n", + "\n", + " # Val pairwise\n", + " val_pair = val_emb[:500] @ val_emb[:500].T\n", + " val_pair_mask = ~torch.eye(500, dtype=torch.bool, device=DEVICE)\n", + " val_pairwise_mean = val_pair[val_pair_mask].mean().item()\n", + "\n", + " # Bank diagnostics on val\n", + " if has_bank:\n", + " _, val_bank_diag = compute_bank_loss(model.bank, val_emb[:2000])\n", + "\n", + " # Enriched diagnostics\n", + " val_enriched_list = []\n", + " for vi in range(0, min(N_VAL, 2000), 512):\n", + " vj = min(vi + 512, min(N_VAL, 2000))\n", + " vo = model(val_ids[vi:vj], val_mask[vi:vj])\n", + " if vo.enriched is not None:\n", + " val_enriched_list.append(vo.enriched)\n", + " if val_enriched_list:\n", + " val_enriched = torch.cat(val_enriched_list)\n", + " geo_ctx = val_enriched[:, 768:]\n", + " geo_ctx_sv = torch.linalg.svdvals(\n", + " geo_ctx.float() - geo_ctx.float().mean(0, keepdim=True))\n", + " geo_eff_dim = float((geo_ctx_sv.sum() ** 2) / (geo_ctx_sv.pow(2).sum() + 1e-12))\n", + " geo_cv = cv_metric(F.normalize(geo_ctx[:1000], dim=-1))\n", + " else:\n", + " val_bank_diag = {}\n", + "\n", + " # ── Epoch tensorboard ──\n", + " writer.add_scalar(\"val/cosine_mean\", val_cos, global_step)\n", + " writer.add_scalar(\"val/cosine_std\", val_cos_std, global_step)\n", + " writer.add_scalar(\"val/retrieval_acc\", val_acc, global_step)\n", + " writer.add_scalar(\"val/cv\", val_cv, global_step)\n", + " writer.add_scalar(\"val/effective_dim\", val_eff_dim, global_step)\n", + " writer.add_scalar(\"val/pairwise_cos_mean\", val_pairwise_mean, global_step)\n", + " writer.add_scalar(\"val/consensus_cv_target\", consensus_cv, global_step)\n", + "\n", + " if val_bank_diag:\n", + " writer.add_scalar(\"val_bank/expert_agreement\", val_bank_diag[\"agreement\"], global_step)\n", + " writer.add_scalar(\"val_bank/rotation_ortho\", val_bank_diag[\"ortho\"], global_step)\n", + " writer.add_scalar(\"val_bank/anchor_spread\", val_bank_diag[\"spread\"], global_step)\n", + " writer.add_scalar(\"val_bank/anchor_entropy\", val_bank_diag[\"entropy\"], global_step)\n", + " writer.add_scalar(\"val_bank/emb_cv\", val_bank_diag[\"emb_cv\"], global_step)\n", + " writer.add_scalar(\"val_bank/expert_cos_mean\", val_bank_diag[\"expert_cos_mean\"], global_step)\n", + " writer.add_scalar(\"val_bank/expert_cos_std\", val_bank_diag[\"expert_cos_std\"], global_step)\n", + " writer.add_scalar(\"val_bank/disagree_preserve\", val_bank_diag[\"disagree\"], global_step)\n", + " writer.add_scalar(\"val_bank/cross_var\", val_bank_diag[\"cross_var\"], global_step)\n", + "\n", + " if val_enriched_list:\n", + " writer.add_scalar(\"val_bank/geo_context_eff_dim\", geo_eff_dim, global_step)\n", + " writer.add_scalar(\"val_bank/geo_context_cv\", geo_cv, global_step)\n", + "\n", + " writer.add_scalar(\"epoch/time_seconds\", elapsed, global_step)\n", + " writer.add_scalar(\"epoch/number\", epoch + 1, global_step)\n", + "\n", + " # ── Console print ──\n", + " print(f\"\\n E{epoch+1:2d}: {elapsed:.0f}s step={global_step}\")\n", + " print(f\" Student: v_cos={val_cos:.4f}±{val_cos_std:.4f} \"\n", + " f\"v_acc={val_acc:.3f} v_cv={val_cv:.4f} eff_dim={val_eff_dim:.1f}\")\n", + " print(f\" Losses: nce={l_nce.item():.4f} mse={l_mse.item():.4f} \"\n", + " f\"bank={l_bank.item() if torch.is_tensor(l_bank) else 0:.4f}\")\n", + " if val_bank_diag:\n", + " print(f\" Bank: agr={val_bank_diag['agreement']:.6f} \"\n", + " f\"ortho={val_bank_diag['ortho']:.6f} \"\n", + " f\"entropy={val_bank_diag['entropy']:.4f} \"\n", + " f\"emb_cv={val_bank_diag['emb_cv']:.4f}\")\n", + " print(f\" exp_cos={val_bank_diag['expert_cos_mean']:.3f}±\"\n", + " f\"{val_bank_diag['expert_cos_std']:.3f} \"\n", + " f\"disagree={val_bank_diag['disagree']:.6f} \"\n", + " f\"spread={val_bank_diag['spread']:.5f}\")\n", + " if val_enriched_list:\n", + " print(f\" Context: geo_eff_dim={geo_eff_dim:.1f} geo_cv={geo_cv:.4f}\")\n", + "\n", + " if val_cos > best_val_cos:\n", + " best_val_cos = val_cos\n", + " torch.save(model.state_dict(), \"cotrain_best.pt\")\n", + " print(f\" ★ New best: v_cos={val_cos:.4f}\")\n", + "\n", + " torch.save(model.state_dict(), \"cotrain_final.pt\")\n", + " writer.close()\n", + "\n", + " # ── Phase 5: Verification ──\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"PHASE 5: VERIFICATION\")\n", + " print(f\"{'='*65}\")\n", + "\n", + " model.load_state_dict(torch.load(\"cotrain_best.pt\", weights_only=True))\n", + " model.eval()\n", + "\n", + " with torch.no_grad():\n", + " val_embs = []\n", + " for vi in range(0, N_VAL, 512):\n", + " vj = min(vi + 512, N_VAL)\n", + " vo = model(val_ids[vi:vj], val_mask[vi:vj])\n", + " val_embs.append(vo.last_hidden_state)\n", + " val_emb = torch.cat(val_embs)\n", + " _, val_acc = infonce(val_emb[:2000], val_targets[:2000])\n", + " val_cos = F.cosine_similarity(val_emb, val_targets, dim=-1).mean().item()\n", + " val_cv = cv_metric(val_emb[:2000])\n", + "\n", + " # Check enriched\n", + " sample_out = model(val_ids[:10], val_mask[:10])\n", + " if sample_out.enriched is not None:\n", + " print(f\" Enriched: {sample_out.enriched.shape}\")\n", + " print(f\" Geo: {sample_out.geometric_context}\")\n", + "\n", + " # Quick similarity test\n", + " test_texts = [\n", + " \"A cat sitting on a windowsill\",\n", + " \"A dog playing in the park\",\n", + " \"A still life painting with flowers\",\n", + " \"A child riding a bicycle\",\n", + " ]\n", + " with torch.no_grad():\n", + " test_in = tokenizer(test_texts, max_length=128, padding=\"max_length\",\n", + " truncation=True, return_tensors=\"pt\").to(DEVICE)\n", + " test_out = model(**test_in)\n", + " test_emb = test_out.last_hidden_state\n", + " sim = test_emb @ test_emb.T\n", + " print(f\"\\n Pairwise cosines:\")\n", + " for i in range(len(test_texts)):\n", + " for j in range(i+1, len(test_texts)):\n", + " print(f\" [{i}]↔[{j}]: {sim[i,j]:.3f} ({test_texts[i][:30]} ↔ {test_texts[j][:30]})\")\n", + "\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"SUMMARY\")\n", + " print(f\"{'='*65}\")\n", + " print(f\" Best v_cos: {best_val_cos:.4f}\")\n", + " print(f\" Final v_cv: {val_cv:.4f}\")\n", + " print(f\" Consensus CV: {consensus_cv:.4f}\")\n", + " print(f\" Val R@1: {val_acc:.3f}\")\n", + " print(f\" Encoder LR: {LR_ENCODER}\")\n", + " print(f\" Bank LR: {LR_BANK}\")\n", + " print(f\" Bank weight: {W_BANK}\")\n", + " print(f\"\\n Saved: cotrain_best.pt, cotrain_final.pt\")\n", + " print(f\" Tensorboard: {log_dir}\")\n", + " print(f\"\\n{'='*65}\")\n", + " print(\"DONE\")\n", + " print(f\"{'='*65}\")\n", + "\n", + "\n", + "if __name__ == \"__main__\":\n", + " run()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000, + "referenced_widgets": [ + "d677f32ed3144b6c89f299e88bdd91db", + "89d9aa5b15c443798086d9b42ce6031e", + "a113c72f2b2c403c991c205293f7baa3", + "b8e7f333003c453eb9d4c85d726f56f5", + "85f4a11c50cc42c9a9e0c9012d72ac04", + "0cd8990f98434e1c9f1afdcf6f54a4e9", + "405998fe651449adaaff88784c1f6c35", + "1c6231d900844c57b2a0e41d3491d824", + "a8d419d176c645c4816b9297f15add8f", + "db04b5a4ac844ee98560ec1ad10e4fbd", + "9ef880c5e59e4bc8b525b782de253c83" + ] + }, + "id": "yjQ1CUaE8BX5", + "outputId": "a7aebd99-389d-44bf-cee4-561f569634bb" + }, + "execution_count": 1, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "=================================================================\n", + "CO-TRAIN: Student + Alignment Bank (unfrozen)\n", + "=================================================================\n", + " Device: cuda\n", + " LR encoder: 0.0001 LR bank: 0.0005\n", + " Bank weight: 0.2\n", + "\n", + "=================================================================\n", + "PHASE 0: LOAD CACHED EMBEDDINGS\n", + "=================================================================\n", + " bert: torch.Size([500000, 768])\n", + " modern: torch.Size([500000, 768])\n", + " roberta: torch.Size([500000, 768])\n", + " albert: torch.Size([500000, 768])\n", + " distil: torch.Size([500000, 768])\n", + " Captions: 500,000, using 500,000\n", + "\n", + "=================================================================\n", + "PHASE 1: GPA ALIGNMENT\n", + "=================================================================\n", + " GPA iter 1: delta=1.99174462\n", + " GPA iter 5: delta=0.00009400\n", + " GPA iter 10: delta=0.00001988\n", + " GPA iter 15: delta=0.00000849\n", + " cos(consensus, bert): 0.9880\n", + " cos(consensus, modern): 0.9831\n", + " cos(consensus, roberta): 0.9885\n", + " cos(consensus, albert): 0.9864\n", + " cos(consensus, distil): 0.9909\n", + " Consensus CV: 0.2543\n", + "\n", + "=================================================================\n", + "PHASE 2: LOAD MODEL (unfrozen)\n", + "=================================================================\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "Loading weights: 0%| | 0/112 [00:00