Hugging Face's logo

dotlabs
/
void.1

Text Generation
Transformers
TensorBoard
Safetensors
English
qwen3
byte-level
pretraining
symbolic
text-generation-inference
Model card Files
xet
 Metrics Community
3
void.1
File size: 377,163 Bytes
b092da0
1
{"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.12.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"accelerator":"GPU","colab":{"gpuType":"T4","provenance":[]},"widgets":{"application/vnd.jupyter.widget-state+json":{"version_major":2,"version_minor":0,"state":{"6c0e573d913a456e824c374c5d80ee49":{"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_5ea71832a1a74fc694eae571e7b0ac0f","IPY_MODEL_7a09d0d404924d7599100af16c415744","IPY_MODEL_21d77936ae5a4e628229303854aaeb32"],"layout":"IPY_MODEL_e1b7dc7e74b441e7a49171ad9f37980b"}},"5ea71832a1a74fc694eae571e7b0ac0f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e3b8db9381564a148ddaee894515fcd5","placeholder":"​","style":"IPY_MODEL_c336016a86024768831f945a9c6d49cb","value":"Fetching 11 files: 100%"}},"7a09d0d404924d7599100af16c415744":{"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_c37e9b6b8f1348dcbacfc94d37a46319","max":11,"min":0,"orientation":"horizontal","style":"IPY_MODEL_bebe9d313b1342778f2d187b94e05458","value":11}},"21d77936ae5a4e628229303854aaeb32":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_a0e7c8482eda43f69b443943316c6f2d","placeholder":"​","style":"IPY_MODEL_17fe5ebd9485444b94ba7bc10299f16b","value":" 11/11 [00:08&lt;00:00,  1.44s/it]"}},"e1b7dc7e74b441e7a49171ad9f37980b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e3b8db9381564a148ddaee894515fcd5":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"c336016a86024768831f945a9c6d49cb":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"c37e9b6b8f1348dcbacfc94d37a46319":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"bebe9d313b1342778f2d187b94e05458":{"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":""}},"a0e7c8482eda43f69b443943316c6f2d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"17fe5ebd9485444b94ba7bc10299f16b":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"d8265d7e0c854cc3a0d5843e5095c426":{"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_c14688e8f19947b08622fecad41255dd","IPY_MODEL_60a4dc787b944932b9017fab086818fd","IPY_MODEL_faedf15e6b654ad6984e43af96aad41f"],"layout":"IPY_MODEL_441cfd7b07a9469e8834af4a15e63dd0"}},"c14688e8f19947b08622fecad41255dd":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_fa02dd7e5ac046a28a26c4f0dbc03787","placeholder":"​","style":"IPY_MODEL_60741011142d4d79b82beafb0750d5c9","value":"README.md: 100%"}},"60a4dc787b944932b9017fab086818fd":{"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_2671696d1afc4a478a60c27a4b2f13dd","max":834,"min":0,"orientation":"horizontal","style":"IPY_MODEL_bc9bbd6412e343bf8761ae96a814aac9","value":834}},"faedf15e6b654ad6984e43af96aad41f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_cca1861469384fa9a1220d0923650861","placeholder":"​","style":"IPY_MODEL_797b0fe55b5044d49641f0bbc17c2498","value":" 834/834 [00:00&lt;00:00, 14.5kB/s]"}},"441cfd7b07a9469e8834af4a15e63dd0":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"fa02dd7e5ac046a28a26c4f0dbc03787":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"60741011142d4d79b82beafb0750d5c9":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"2671696d1afc4a478a60c27a4b2f13dd":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"bc9bbd6412e343bf8761ae96a814aac9":{"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":""}},"cca1861469384fa9a1220d0923650861":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"797b0fe55b5044d49641f0bbc17c2498":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"2ab55c29107c4d8b9653209d91983755":{"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_1121ac17c42140bcaed0970454800b4a","IPY_MODEL_9c388a6c25ea47c49b2b22dd24b7e445","IPY_MODEL_0e9b66113253434f9d23617b1f5d6119"],"layout":"IPY_MODEL_dfa95837b9b445319294c3db64231abf"}},"1121ac17c42140bcaed0970454800b4a":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_76c668899a984e38bffcd2f69c3ce14b","placeholder":"​","style":"IPY_MODEL_8a182f6ba34e4aa08d5275775f62b443","value":"config.json: "}},"9c388a6c25ea47c49b2b22dd24b7e445":{"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_0710dbab770342a0815e25dac044eee3","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_1ccb953779ea497fbef3f5a5b3058a5c","value":1}},"0e9b66113253434f9d23617b1f5d6119":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_d505968ea62246f39d9f0382f9aa0953","placeholder":"​","style":"IPY_MODEL_39389636f31b458186e2795290f4c42b","value":" 1.28k/? [00:00&lt;00:00, 13.5kB/s]"}},"dfa95837b9b445319294c3db64231abf":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"76c668899a984e38bffcd2f69c3ce14b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"8a182f6ba34e4aa08d5275775f62b443":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0710dbab770342a0815e25dac044eee3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"1ccb953779ea497fbef3f5a5b3058a5c":{"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":""}},"d505968ea62246f39d9f0382f9aa0953":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"39389636f31b458186e2795290f4c42b":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"df962532bf794d88a5d3457cf311b6f5":{"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_850c24c1a9284b65917c756b72d60f91","IPY_MODEL_15e39e3a50c747cfa45b618a3e7403b4","IPY_MODEL_31a1528334bc49108cf33608e0512359"],"layout":"IPY_MODEL_b87048f0c61e425591d559f5c214de93"}},"850c24c1a9284b65917c756b72d60f91":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_b355333668b74860a2f81da3d8048af8","placeholder":"​","style":"IPY_MODEL_a03715fd2cda406b93acd98f5c6c41af","value":"byte_vocab.json: "}},"15e39e3a50c747cfa45b618a3e7403b4":{"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_4a7266d0b24a4108868a6711256127c7","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_7e63a22f6c684f91bd656c3e65da6158","value":1}},"31a1528334bc49108cf33608e0512359":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_54f2446d49c346ce834de1befa18bd2c","placeholder":"​","style":"IPY_MODEL_4c1d62f9fc554466947c9973963495ff","value":" 4.29k/? [00:00&lt;00:00, 72.5kB/s]"}},"b87048f0c61e425591d559f5c214de93":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"b355333668b74860a2f81da3d8048af8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a03715fd2cda406b93acd98f5c6c41af":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"4a7266d0b24a4108868a6711256127c7":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"7e63a22f6c684f91bd656c3e65da6158":{"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":""}},"54f2446d49c346ce834de1befa18bd2c":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"4c1d62f9fc554466947c9973963495ff":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"081b6e87488d46c28c7462d66616ced6":{"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_8bbf6e08ec5d458fb20ee784300b2c87","IPY_MODEL_e171e8b6a53f408fb0b8d5ae9b7609dc","IPY_MODEL_649b37e2e92e40f4b4ac25dbbfad5f3f"],"layout":"IPY_MODEL_65ec99f55e1b4fcb817e7e44b5a067b3"}},"8bbf6e08ec5d458fb20ee784300b2c87":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_a7d0728325d04035b20696a66535cb46","placeholder":"​","style":"IPY_MODEL_752ee572888e40899bc7b836b79a5d58","value":"model.safetensors: 100%"}},"e171e8b6a53f408fb0b8d5ae9b7609dc":{"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_841ef6d222b14448bf7b622a1c4da3b8","max":360622744,"min":0,"orientation":"horizontal","style":"IPY_MODEL_49fd4cd23f6e4a89986aca01a3392a73","value":360622744}},"649b37e2e92e40f4b4ac25dbbfad5f3f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_4a4eadf90877428db3d75dd2edc0e466","placeholder":"​","style":"IPY_MODEL_68ec2ac98ca64edbb992163da859c3d8","value":" 361M/361M [00:07&lt;00:00, 46.8MB/s]"}},"65ec99f55e1b4fcb817e7e44b5a067b3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a7d0728325d04035b20696a66535cb46":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"752ee572888e40899bc7b836b79a5d58":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"841ef6d222b14448bf7b622a1c4da3b8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"49fd4cd23f6e4a89986aca01a3392a73":{"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":""}},"4a4eadf90877428db3d75dd2edc0e466":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"68ec2ac98ca64edbb992163da859c3d8":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0140b5c10c804ae4a654ff5b46139589":{"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_038709664d8741fca713c83e00b83966","IPY_MODEL_1fa509840f8941bda369d75381130e26","IPY_MODEL_483f08aa35a34726ad3225112935542a"],"layout":"IPY_MODEL_5967ab3622fd49399135dd9f4d9a35b5"}},"038709664d8741fca713c83e00b83966":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f46906bfd6a24951b44aa28a1a2b07b6","placeholder":"​","style":"IPY_MODEL_bdb7c09db6a04dce993e15167ee9c547","value":"latest.json: "}},"1fa509840f8941bda369d75381130e26":{"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_48a87645bef44cff906c6f3e801cea72","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_cd41166b07b64624af1ad5b0cad1f32a","value":1}},"483f08aa35a34726ad3225112935542a":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_0eff1c19e8e14e778a9fb52b48aabd01","placeholder":"​","style":"IPY_MODEL_76f3aee8199945408afe5bcad7159f15","value":" 1.20k/? [00:00&lt;00:00, 20.5kB/s]"}},"5967ab3622fd49399135dd9f4d9a35b5":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"f46906bfd6a24951b44aa28a1a2b07b6":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"bdb7c09db6a04dce993e15167ee9c547":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"48a87645bef44cff906c6f3e801cea72":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"cd41166b07b64624af1ad5b0cad1f32a":{"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":""}},"0eff1c19e8e14e778a9fb52b48aabd01":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"76f3aee8199945408afe5bcad7159f15":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"c50428db770e4c5abde921a32badde1d":{"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_d7889da7c8404b6e8cd3f701eeb0c2b4","IPY_MODEL_d09e05d7057e421caddca02aaaa0b666","IPY_MODEL_314a15413f874f7a8aefa0dabb8e5508"],"layout":"IPY_MODEL_ab54a4e50539487685869e29a24e7402"}},"d7889da7c8404b6e8cd3f701eeb0c2b4":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f2d47917e3bb4000bd09bd24543191a3","placeholder":"​","style":"IPY_MODEL_d8d489fd384c45db93a2fc2b3f12347b","value":"special_tokens_map.json: 100%"}},"d09e05d7057e421caddca02aaaa0b666":{"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_7800df79b87b46a5af9c1489e690a177","max":75,"min":0,"orientation":"horizontal","style":"IPY_MODEL_41196335ecf148c5a8bc3c1a54dffb9b","value":75}},"314a15413f874f7a8aefa0dabb8e5508":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_afc72f9285bf4fee9fadb058f3201290","placeholder":"​","style":"IPY_MODEL_7e4672c8555849b98267fb4c3a491710","value":" 75.0/75.0 [00:00&lt;00:00, 1.56kB/s]"}},"ab54a4e50539487685869e29a24e7402":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"f2d47917e3bb4000bd09bd24543191a3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"d8d489fd384c45db93a2fc2b3f12347b":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"7800df79b87b46a5af9c1489e690a177":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"41196335ecf148c5a8bc3c1a54dffb9b":{"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":""}},"afc72f9285bf4fee9fadb058f3201290":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7e4672c8555849b98267fb4c3a491710":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"fd53ada9a1f84724b0384ffb7fe66732":{"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_a8bcf85fcb8f4da885d7ec433588ad4c","IPY_MODEL_31349f8606db4827bdc9a7d87698f7a3","IPY_MODEL_51c801bf4d1f40339e96b15e569411dd"],"layout":"IPY_MODEL_eacffd8c09b448549f29d942ab0c9061"}},"a8bcf85fcb8f4da885d7ec433588ad4c":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_9d1f385e5dae46f59b78e748a7280a4f","placeholder":"​","style":"IPY_MODEL_7ff9020b41e544eab4f86c2e45a8d092","value":"generation_config.json: 100%"}},"31349f8606db4827bdc9a7d87698f7a3":{"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_7492fb9555d44b5b9d847526ba284490","max":160,"min":0,"orientation":"horizontal","style":"IPY_MODEL_70b064e3df314bcdae4502e5a67a7c28","value":160}},"51c801bf4d1f40339e96b15e569411dd":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_4cfee8f819624c95a3d74ba9558b64b7","placeholder":"​","style":"IPY_MODEL_478368ea1ccd43d8830d0f75fc67e2a1","value":" 160/160 [00:00&lt;00:00, 2.73kB/s]"}},"eacffd8c09b448549f29d942ab0c9061":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9d1f385e5dae46f59b78e748a7280a4f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7ff9020b41e544eab4f86c2e45a8d092":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"7492fb9555d44b5b9d847526ba284490":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"70b064e3df314bcdae4502e5a67a7c28":{"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":""}},"4cfee8f819624c95a3d74ba9558b64b7":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"478368ea1ccd43d8830d0f75fc67e2a1":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0574c046687f4eada2317639551183b8":{"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_434dfeda872f4ae290d960c89104a486","IPY_MODEL_9b4161a96f21490b8c858602f65844a6","IPY_MODEL_bc0eff8590fa48719d3960164c641e79"],"layout":"IPY_MODEL_157087b33015475a8226b4c86aed416d"}},"434dfeda872f4ae290d960c89104a486":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_229fb0b6bb0c4407b30a06641776df57","placeholder":"​","style":"IPY_MODEL_a48b779ebb864f3a8955ff279b2f44bd","value":"coverage.json: "}},"9b4161a96f21490b8c858602f65844a6":{"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_db0a2dd0d4e94cfe91deaf41901ee8d3","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_7e2b25e3ff3e4c4da633e9b0de58922b","value":1}},"bc0eff8590fa48719d3960164c641e79":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_518fefd15e414aba908e29b6842a4e90","placeholder":"​","style":"IPY_MODEL_046639e8901e42c280a874f728c9d03a","value":" 2.27k/? [00:00&lt;00:00, 31.3kB/s]"}},"157087b33015475a8226b4c86aed416d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"229fb0b6bb0c4407b30a06641776df57":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a48b779ebb864f3a8955ff279b2f44bd":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"db0a2dd0d4e94cfe91deaf41901ee8d3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"7e2b25e3ff3e4c4da633e9b0de58922b":{"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":""}},"518fefd15e414aba908e29b6842a4e90":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"046639e8901e42c280a874f728c9d03a":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"1f2de106078446218506020442c7aeaf":{"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_54d6972637724629bf393072c086d813","IPY_MODEL_f3329ed867054545b3a045af9c75de05","IPY_MODEL_1ae970fd760d47978f3af86038f26f0f"],"layout":"IPY_MODEL_7edc99ebc93e436aadcd7b813f78d92c"}},"54d6972637724629bf393072c086d813":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_c811209d81e34863969e6f4f0a2c2877","placeholder":"​","style":"IPY_MODEL_a81608e6abd74f7387ee004a8e6977af","value":"tokenization_byte.py: "}},"f3329ed867054545b3a045af9c75de05":{"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_9607ab5565dd486f87241b1417b294c8","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_9614bf63dd2e4d4c9802a74f113d5c14","value":1}},"1ae970fd760d47978f3af86038f26f0f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_aa4b5dac1cbd4d0aa5ed74b367467c3d","placeholder":"​","style":"IPY_MODEL_3b43a8c3c26c4d2490237f7ac0aa9d61","value":" 4.46k/? [00:00&lt;00:00, 239kB/s]"}},"7edc99ebc93e436aadcd7b813f78d92c":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"c811209d81e34863969e6f4f0a2c2877":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a81608e6abd74f7387ee004a8e6977af":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"9607ab5565dd486f87241b1417b294c8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"9614bf63dd2e4d4c9802a74f113d5c14":{"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":""}},"aa4b5dac1cbd4d0aa5ed74b367467c3d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"3b43a8c3c26c4d2490237f7ac0aa9d61":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0b590d19f70944e0960d539695a49c3f":{"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_f84dd6047ded477e8bdea1495560bc8c","IPY_MODEL_cc9402b5f52341e58d09928c67e17d21","IPY_MODEL_a91a15d761704f908840a9ac48e69543"],"layout":"IPY_MODEL_b03df882d8424bbdbbb00857d82a79a2"}},"f84dd6047ded477e8bdea1495560bc8c":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f2073ca8c1c6467cad9ca2402d16c575","placeholder":"​","style":"IPY_MODEL_aaf89a5a64a54701819bb8e2e2984bb6","value":"training_config.json: "}},"cc9402b5f52341e58d09928c67e17d21":{"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_8dd773d1eefd469cae10f3bd3e9e627a","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_18d09157b6e947929301a0855e87e71c","value":1}},"a91a15d761704f908840a9ac48e69543":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_d2c5844fae45496a9a6be73d7d1fd484","placeholder":"​","style":"IPY_MODEL_9d9bb1572c4746c7a294f2fc87bb9702","value":" 1.68k/? [00:00&lt;00:00, 80.4kB/s]"}},"b03df882d8424bbdbbb00857d82a79a2":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"f2073ca8c1c6467cad9ca2402d16c575":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"aaf89a5a64a54701819bb8e2e2984bb6":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"8dd773d1eefd469cae10f3bd3e9e627a":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"18d09157b6e947929301a0855e87e71c":{"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":""}},"d2c5844fae45496a9a6be73d7d1fd484":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9d9bb1572c4746c7a294f2fc87bb9702":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"4255d387aac448ab8dad7a3d2b5e7a52":{"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_decc4490cb19434e8f62398b60bea037","IPY_MODEL_00776136935f4d7e912c4ea7b18422a9","IPY_MODEL_1f5cf831a673493ab7e75334826dd039"],"layout":"IPY_MODEL_4045711b6e804a70bdf635c260c96f32"}},"decc4490cb19434e8f62398b60bea037":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_d6e0199799634196972c73eba67946f6","placeholder":"​","style":"IPY_MODEL_59bb7e0151914493b281a805ac80fda0","value":"tokenizer_config.json: 100%"}},"00776136935f4d7e912c4ea7b18422a9":{"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_b583d6dbaa8f4ce9b8cdabc5ff9fd78a","max":932,"min":0,"orientation":"horizontal","style":"IPY_MODEL_56e8a5d918b74c02ae520a9d0e23457b","value":932}},"1f5cf831a673493ab7e75334826dd039":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_ddf6f88c0bef47058d474c326a90e53e","placeholder":"​","style":"IPY_MODEL_0fa75cf0dcae426ab3b4ccff7e8017ac","value":" 932/932 [00:00&lt;00:00, 21.9kB/s]"}},"4045711b6e804a70bdf635c260c96f32":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"d6e0199799634196972c73eba67946f6":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"59bb7e0151914493b281a805ac80fda0":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b583d6dbaa8f4ce9b8cdabc5ff9fd78a":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"56e8a5d918b74c02ae520a9d0e23457b":{"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":""}},"ddf6f88c0bef47058d474c326a90e53e":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"0fa75cf0dcae426ab3b4ccff7e8017ac":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"454a64c1b29c416fbd7be768488027b3":{"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_19f35d68a8aa4c679ff9bbcc8d145f0e","IPY_MODEL_20eebefa4a5f4c1782e41d20a5199b76","IPY_MODEL_f5a146767d5446ec9a62888d02c5d3df"],"layout":"IPY_MODEL_bff1824dfcb1442d83c30440b39764dd"}},"19f35d68a8aa4c679ff9bbcc8d145f0e":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_b575e4632508471398023e563cf1a5d0","placeholder":"​","style":"IPY_MODEL_2953c3a16a634ea6a22fe78ef706ab57","value":"benchmark.py: "}},"20eebefa4a5f4c1782e41d20a5199b76":{"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_167e7db8138c4dca9037528709524a49","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_5f4d2371084748f8aa1e988fec6c09f6","value":1}},"f5a146767d5446ec9a62888d02c5d3df":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_8fcd4fcf166641deaf192294feed096d","placeholder":"​","style":"IPY_MODEL_10dcccd3ba18474cb6947da57da155d0","value":" 33.3k/? [00:00&lt;00:00, 2.20MB/s]"}},"bff1824dfcb1442d83c30440b39764dd":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"b575e4632508471398023e563cf1a5d0":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"2953c3a16a634ea6a22fe78ef706ab57":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"167e7db8138c4dca9037528709524a49":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"5f4d2371084748f8aa1e988fec6c09f6":{"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":""}},"8fcd4fcf166641deaf192294feed096d":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"10dcccd3ba18474cb6947da57da155d0":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"6decf21dd8b6432e8e9119776aaecd4f":{"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_770bad2022af4be88752ae58f73a688b","IPY_MODEL_b1be3ee818b04e2ca6544069e75dcb61","IPY_MODEL_a8997d8ee6f94441803a3c171a1524cc"],"layout":"IPY_MODEL_22d5d5dfbe824bb491d9c98d234b00a6"}},"770bad2022af4be88752ae58f73a688b":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f88ee16075ab4d03bc9376ab6d5762d1","placeholder":"​","style":"IPY_MODEL_2ab67ac614404fbc865e7c396f8ac5d7","value":"bencharithmark-3.py: "}},"b1be3ee818b04e2ca6544069e75dcb61":{"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_c4a9a778f9eb4f5c84747d1f722c8bd0","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_da0d938e11e8450ab0a1f6efcfbb1c5b","value":1}},"a8997d8ee6f94441803a3c171a1524cc":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_13b1fd360be24a3a93d73e30d6f24033","placeholder":"​","style":"IPY_MODEL_b1a21592565b4430926b7e9217d583d2","value":" 24.7k/? [00:00&lt;00:00, 2.00MB/s]"}},"22d5d5dfbe824bb491d9c98d234b00a6":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"f88ee16075ab4d03bc9376ab6d5762d1":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"2ab67ac614404fbc865e7c396f8ac5d7":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"c4a9a778f9eb4f5c84747d1f722c8bd0":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"da0d938e11e8450ab0a1f6efcfbb1c5b":{"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":""}},"13b1fd360be24a3a93d73e30d6f24033":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"b1a21592565b4430926b7e9217d583d2":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"7652e9089b104c9d91988b49e30ab748":{"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_5506dc21fc2c4d7896d45c9b75830293","IPY_MODEL_db79212074c4426fa86c1d885be4b341","IPY_MODEL_73bfc6154a5d449cbfd57dae9cab705c"],"layout":"IPY_MODEL_cbfe3fdb0f884880ba1f080ce4232d09"}},"5506dc21fc2c4d7896d45c9b75830293":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e9a4d884660048c8bd8321563f0ff855","placeholder":"​","style":"IPY_MODEL_fbdc5adcc5e349df9978d7877956d997","value":"arithmark-3.jsonl: "}},"db79212074c4426fa86c1d885be4b341":{"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_b4a983f8062044328c595abfeef09e6f","max":1,"min":0,"orientation":"horizontal","style":"IPY_MODEL_56b3d94a40514bf6b7ab6369bae6be71","value":1}},"73bfc6154a5d449cbfd57dae9cab705c":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_f538ca606f754965be6359213171709b","placeholder":"​","style":"IPY_MODEL_5d34eeb0d6b94968be6fa15102cd0e80","value":" 787k/? [00:00&lt;00:00, 42.9MB/s]"}},"cbfe3fdb0f884880ba1f080ce4232d09":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e9a4d884660048c8bd8321563f0ff855":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"fbdc5adcc5e349df9978d7877956d997":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b4a983f8062044328c595abfeef09e6f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_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"}},"56b3d94a40514bf6b7ab6369bae6be71":{"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":""}},"f538ca606f754965be6359213171709b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"5d34eeb0d6b94968be6fa15102cd0e80":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}}}}}},"nbformat_minor":5,"nbformat":4,"cells":[{"id":"63d51cef","cell_type":"markdown","source":"# Void — Inference + Benchmark Notebook\n\nThis notebook benchmarks **`appvoid/void-byte`** using the same evaluation family used for the looped Fig model, but removes every recurrent/loop-specific mechanism.\n\nVoid is a standard decoder-only **Qwen3 causal LM** with:\n\n- **90,148,352 parameters**\n- **24 independent transformer layers**\n- **hidden size 640**\n- **MLP 1440**\n- **10 query heads / 2 KV heads**\n- **64-dimensional attention heads**\n- **259-token byte vocabulary**: bytes `0..255`, PAD `256`, BOS `257`, EOS `258`\n- **native context 2048**\n\nThere are **no cycles, recurrent modes, `BET_EVAL_CYCLES`, loop-routing tensors, or L10 export conversion**.\n\n## Benchmarks\n\nThe notebook keeps the same benchmark suite and scoring structure:\n\n- HellaSwag\n- PIQA\n- ARC Easy\n- ARC Challenge\n- BananaMind Base Bench 1.1\n- ArithMark 3.0\n- the same chance-normalized **Intelligence Index**\n\nFor apples-to-apples comparison with the earlier Fig benchmark, `BENCH_MAX_LENGTH` defaults to **1024**. Void itself supports 2048; change that one setting if you want a native-context run instead.\n","metadata":{"id":"63d51cef"}},{"id":"1a19c1b2","cell_type":"markdown","source":"## 1. Install dependencies","metadata":{"id":"1a19c1b2"}},{"id":"c379b06c","cell_type":"code","source":"import sys, subprocess, os\n\nPACKAGES = [\n    \"transformers>=4.56,<5\",\n    \"huggingface_hub>=0.34\",\n    \"datasets>=3.0\",\n    \"accelerate>=1.0\",\n    \"safetensors>=0.4\",\n    \"lm-eval[hf]==0.4.12\",\n    \"pandas>=2.0\",\n    \"matplotlib>=3.8\",\n    \"ipywidgets>=8.1\",\n]\n\nsubprocess.check_call(\n    [sys.executable, \"-m\", \"pip\", \"install\", \"-q\", \"--upgrade\", *PACKAGES]\n)\n\nprint(\"Installed with:\", sys.executable)\n","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"c379b06c","outputId":"921822e1-ca36-4725-a924-7a6e4f96f62b","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:13:51.073081Z","iopub.execute_input":"2026-10-09T02:13:51.073394Z","iopub.status.idle":"2026-10-09T02:13:57.552188Z","shell.execute_reply.started":"2026-10-09T02:13:51.073367Z","shell.execute_reply":"2026-10-09T02:13:57.551359Z"}},"outputs":[{"name":"stdout","text":"Installed with: /usr/bin/python3\n","output_type":"stream"}],"execution_count":7},{"id":"2a6ee67b","cell_type":"markdown","source":"## 2. Optional Hugging Face authentication\n\nPublic repositories need no token: leave `USE_HF_TOKEN = False`. No login or token prompt is used. For private/gated repositories, set it to `True` and supply `HF_TOKEN` through your environment or Kaggle/Colab secrets; never paste a token into this notebook.\n","metadata":{"id":"2a6ee67b"}},{"id":"d40a6fc9","cell_type":"code","source":"import os\n\nUSE_HF_TOKEN = True  # Public repository: no token needed.\n\ndef acquire_hf_token():\n    if not USE_HF_TOKEN:\n        return False  # Explicit anonymous Hub requests, even with a cached login.\n    token = os.environ.get(\"HF_TOKEN\") or os.environ.get(\"HUGGING_FACE_HUB_TOKEN\")\n    if not token:\n        try:\n            from kaggle_secrets import UserSecretsClient\n            token = UserSecretsClient().get_secret(\"HF_TOKEN\")\n        except Exception:\n            pass\n    if not token:\n        try:\n            from google.colab import userdata\n            token = userdata.get(\"HF_TOKEN\")\n        except Exception:\n            pass\n    if not token or not token.strip():\n        raise RuntimeError(\"Private access enabled: set HF_TOKEN in environment/secrets, or use USE_HF_TOKEN = False for a public repo.\")\n    return token.strip()\n\nHF_TOKEN = acquire_hf_token()\nos.environ[\"HF_HUB_DISABLE_IMPLICIT_TOKEN\"] = \"1\"\nif HF_TOKEN:\n    os.environ[\"HF_TOKEN\"] = HF_TOKEN\nelse:\n    for key in (\"HF_TOKEN\", \"HUGGING_FACE_HUB_TOKEN\"):\n        os.environ.pop(key, None)\n\n# Keep reruns consistent if huggingface_hub was imported in this kernel already.\nfrom huggingface_hub import constants as hf_constants\nhf_constants.HF_HUB_DISABLE_IMPLICIT_TOKEN = True\nprint(\"Hugging Face access:\", \"optional token enabled\" if HF_TOKEN else \"anonymous (no token)\")\n","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"d40a6fc9","outputId":"d4c148de-53ed-4055-cd2a-68983e4279e9","trusted":true,"execution":{"iopub.status.busy":"2026-10-09T02:13:57.553696Z","iopub.execute_input":"2026-10-09T02:13:57.554017Z","iopub.status.idle":"2026-10-09T02:13:57.561276Z","shell.execute_reply.started":"2026-10-09T02:13:57.553994Z","shell.execute_reply":"2026-10-09T02:13:57.560351Z"}},"outputs":[{"name":"stdout","text":"Hugging Face access: optional token enabled\n","output_type":"stream"}],"execution_count":8},{"id":"5910a258","cell_type":"markdown","source":"## 3. Choose a checkpoint\n\nSet `MODEL_REVISION` to any commit SHA, tag, or branch (`\"main\"` for latest). The default below is your supplied **step 900,000** revision. Each resolved commit gets its own model and results directory. After changing the revision, rerun from this configuration cell downward (or restart and Run All).\n","metadata":{"id":"5910a258"}},{"id":"aee0ecb4","cell_type":"code","source":"from pathlib import Path\nimport torch, json, time, gc, math, shutil, subprocess, sys, os\nimport pandas as pd\n\nMODEL_ID = \"dotlabs/void.1\"\n# Step 900,000 (user-provided checkpoint mapping).\nMODEL_REVISION = \"92c601560cc2179e3bfc33e067fdd3bf9254e163\"\nMODEL_REVISION = MODEL_REVISION.strip()\nif not MODEL_REVISION:\n    raise ValueError(\"MODEL_REVISION must be a commit SHA, tag, or branch.\")\n\nEXPECTED_PARAMS = 90_148_352\nEXPECTED_VOCAB = 259\n\nDEVICE = \"cuda:0\" if torch.cuda.is_available() else \"cpu\"\n\n# Match the earlier Fig benchmark protocol by default.\n# Void can also run BF16 on supported GPUs; switch both values if you want that variant.\nif torch.cuda.is_available():\n    TORCH_DTYPE = torch.float16\n    BENCH_DTYPE = \"float16\"\nelse:\n    TORCH_DTYPE = torch.float32\n    BENCH_DTYPE = \"float32\"\n\n# Keep 1024 by default to match the earlier Fig benchmark protocol.\n# Set to 2048 for a native-context Void run.\nBENCH_MAX_LENGTH = 2048\n\nRESULTS_BASE = (\n    Path(\"/kaggle/working/void_eval\")\n    if Path(\"/kaggle/working\").exists()\n    else Path.cwd() / \"void_eval\"\n)\nROOT = RESULTS_BASE\nROOT.mkdir(parents=True, exist_ok=True)\n\nLOCAL_MODEL_DIR = (ROOT / \"_void_model\").resolve()\nACTIVE_MODEL = None\nRESOLVED_MODEL_REVISION = None\n\nprint(\"Model:\", MODEL_ID)\nprint(\"Requested revision:\", MODEL_REVISION)\nprint(\"Device:\", DEVICE)\nprint(\"Benchmark dtype:\", BENCH_DTYPE)\nprint(\"Benchmark max length:\", BENCH_MAX_LENGTH)\nprint(\"Results directory:\", ROOT)\n","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"aee0ecb4","outputId":"1fa4b863-1e1c-44ae-cf6c-8f014ac3825e","trusted":true,"execution":{"iopub.status.busy":"2026-10-09T02:13:57.562080Z","iopub.execute_input":"2026-10-09T02:13:57.562757Z","iopub.status.idle":"2026-10-09T02:13:57.587246Z","shell.execute_reply.started":"2026-10-09T02:13:57.562720Z","shell.execute_reply":"2026-10-09T02:13:57.586323Z"}},"outputs":[{"name":"stdout","text":"Model: appvoid/void-byte\nRequested revision: 92c601560cc2179e3bfc33e067fdd3bf9254e163\nDevice: cuda:0\nBenchmark dtype: float16\nBenchmark max length: 2048\nResults directory: /kaggle/working/void_eval\n","output_type":"stream"}],"execution_count":9},{"id":"f29aaabc","cell_type":"markdown","source":"## 4. Load the selected Void checkpoint\n\nThe benchmark pins one Hub snapshot locally and all child benchmark processes use that exact snapshot. This avoids comparing different checkpoints if the repository changes while a long suite is running.\n\nUnlike Fig, this is a normal Qwen3 checkpoint: there is no model surgery or recurrent-depth export.\n","metadata":{"id":"f29aaabc"}},{"id":"c001d0d1","cell_type":"code","source":"from transformers import AutoTokenizer, AutoModelForCausalLM\nfrom huggingface_hub import snapshot_download, HfApi\n\n# Release a previously loaded checkpoint when rerunning this cell.\nif globals().get(\"void_model\") is not None:\n    unload_void()\nvoid_tokenizer = None\nvoid_model = None\n\ndef prepare_model_snapshot(force_refresh=False):\n    global ACTIVE_MODEL, RESOLVED_MODEL_REVISION, LOCAL_MODEL_DIR, ROOT\n\n    # Clear previous state first: a failed lookup must never reuse another checkpoint.\n    if void_model is not None:\n        unload_void()\n    ACTIVE_MODEL = None\n    RESOLVED_MODEL_REVISION = None\n    api = HfApi()\n    try:\n        info = api.model_info(MODEL_ID, revision=MODEL_REVISION, token=HF_TOKEN)\n    except Exception as exc:\n        raise RuntimeError(\n            f\"Cannot access {MODEL_ID} at revision {MODEL_REVISION!r}. \"\n            \"Check that the revision exists and the repo is public, or enable optional authentication for private access.\"\n        ) from exc\n    resolved_revision = info.sha\n    if not resolved_revision:\n        raise RuntimeError(\"Hub did not return a resolved commit SHA.\")\n    ROOT = RESULTS_BASE / MODEL_ID.replace(\"/\", \"--\") / resolved_revision\n    ROOT.mkdir(parents=True, exist_ok=True)\n    LOCAL_MODEL_DIR = (ROOT / \"_void_model\").resolve()\n    if force_refresh and LOCAL_MODEL_DIR.exists():\n        shutil.rmtree(LOCAL_MODEL_DIR)\n\n    # Only materialize files needed for inference/benchmarking plus checkpoint metadata.\n    patterns = [\n        \"config.json\",\n        \"generation_config.json\",\n        \"model.safetensors\",\n        \"model-*.safetensors\",\n        \"model.safetensors.index.json\",\n        \"pytorch_model.bin\",\n        \"pytorch_model-*.bin\",\n        \"pytorch_model.bin.index.json\",\n        \"tokenizer_config.json\",\n        \"special_tokens_map.json\",\n        \"byte_vocab.json\",\n        \"tokenization_byte.py\",\n        \"latest.json\",\n        \"training_config.json\",\n        \"coverage.json\",\n        \"README.md\",\n    ]\n\n    path = Path(snapshot_download(\n        repo_id=MODEL_ID,\n        revision=resolved_revision,\n        token=HF_TOKEN,\n        local_dir=str(LOCAL_MODEL_DIR),\n        allow_patterns=patterns,\n    )).resolve()\n\n    config_path = path / \"config.json\"\n    if not config_path.exists():\n        raise FileNotFoundError(f\"Expected model config at {config_path}\")\n\n    cfg = json.loads(config_path.read_text(encoding=\"utf-8\"))\n    RESOLVED_MODEL_REVISION = resolved_revision\n    ACTIVE_MODEL = str(path)\n\n    print(\"Pinned model snapshot:\", ACTIVE_MODEL)\n    print(\"Resolved Hub revision:\", RESOLVED_MODEL_REVISION)\n    print(\"Architecture:\", cfg.get(\"architectures\", [\"unknown\"])[0])\n    print(\"Model type:\", cfg.get(\"model_type\", \"unknown\"))\n    print(\"Layers:\", cfg.get(\"num_hidden_layers\", \"unknown\"))\n    print(\"Hidden size:\", cfg.get(\"hidden_size\", \"unknown\"))\n    print(\"MLP size:\", cfg.get(\"intermediate_size\", \"unknown\"))\n    print(\"Attention heads:\", cfg.get(\"num_attention_heads\", \"unknown\"))\n    print(\"KV heads:\", cfg.get(\"num_key_value_heads\", \"unknown\"))\n    print(\"Head dim:\", cfg.get(\"head_dim\", \"unknown\"))\n    print(\"Vocab:\", cfg.get(\"vocab_size\", \"unknown\"))\n    print(\"Native context:\", cfg.get(\"max_position_embeddings\", \"unknown\"))\n\n    latest = path / \"latest.json\"\n    if latest.exists():\n        try:\n            latest_meta = json.loads(latest.read_text(encoding=\"utf-8\"))\n            print(\"Checkpoint step:\", latest_meta.get(\"step\", \"unknown\"))\n        except Exception:\n            pass\n\n    return ACTIVE_MODEL\n\n\ndef load_void(force_reload=False):\n    global void_tokenizer, void_model\n\n    if ACTIVE_MODEL is None:\n        prepare_model_snapshot()\n\n    if void_model is not None and not force_reload:\n        return void_model, void_tokenizer\n\n    if force_reload:\n        unload_void()\n\n    print(f\"Loading Void from pinned snapshot {ACTIVE_MODEL} ...\")\n\n    void_tokenizer = AutoTokenizer.from_pretrained(\n        ACTIVE_MODEL,\n        token=HF_TOKEN,\n        trust_remote_code=True,\n        use_fast=False,\n    )\n\n    kwargs = dict(\n        token=HF_TOKEN,\n        trust_remote_code=True,\n        low_cpu_mem_usage=True,\n        attn_implementation=\"sdpa\",\n    )\n    try:\n        void_model = AutoModelForCausalLM.from_pretrained(\n            ACTIVE_MODEL,\n            dtype=TORCH_DTYPE,\n            **kwargs,\n        )\n    except TypeError:\n        # Compatibility with Transformers versions that still use torch_dtype.\n        void_model = AutoModelForCausalLM.from_pretrained(\n            ACTIVE_MODEL,\n            torch_dtype=TORCH_DTYPE,\n            **kwargs,\n        )\n\n    void_model = void_model.to(DEVICE).eval()\n\n    params = sum(p.numel() for p in void_model.parameters())\n    cfg = void_model.config\n\n    print(f\"Loaded: {params:,} parameters\")\n    print(\"Context:\", getattr(cfg, \"max_position_embeddings\", \"unknown\"))\n    print(\"Vocab:\", getattr(cfg, \"vocab_size\", \"unknown\"))\n    print(\"BOS/PAD/EOS:\",\n          void_tokenizer.bos_token_id,\n          void_tokenizer.pad_token_id,\n          void_tokenizer.eos_token_id)\n\n    if params != EXPECTED_PARAMS:\n        print(f\"WARNING: expected {EXPECTED_PARAMS:,} parameters, found {params:,}.\")\n    if int(getattr(cfg, \"vocab_size\", -1)) != EXPECTED_VOCAB:\n        raise RuntimeError(\n            f\"Void benchmark expects the 259-token byte model, \"\n            f\"but config.vocab_size={getattr(cfg, 'vocab_size', None)}.\"\n        )\n\n    return void_model, void_tokenizer\n\n\ndef unload_void():\n    global void_tokenizer, void_model\n    void_model = None\n    void_tokenizer = None\n    gc.collect()\n    if torch.cuda.is_available():\n        torch.cuda.empty_cache()\n\n\nprepare_model_snapshot()\nvoid_model, void_tokenizer = load_void()\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":713,"referenced_widgets":["6c0e573d913a456e824c374c5d80ee49","5ea71832a1a74fc694eae571e7b0ac0f","7a09d0d404924d7599100af16c415744","21d77936ae5a4e628229303854aaeb32","e1b7dc7e74b441e7a49171ad9f37980b","e3b8db9381564a148ddaee894515fcd5","c336016a86024768831f945a9c6d49cb","c37e9b6b8f1348dcbacfc94d37a46319","bebe9d313b1342778f2d187b94e05458","a0e7c8482eda43f69b443943316c6f2d","17fe5ebd9485444b94ba7bc10299f16b","d8265d7e0c854cc3a0d5843e5095c426","c14688e8f19947b08622fecad41255dd","60a4dc787b944932b9017fab086818fd","faedf15e6b654ad6984e43af96aad41f","441cfd7b07a9469e8834af4a15e63dd0","fa02dd7e5ac046a28a26c4f0dbc03787","60741011142d4d79b82beafb0750d5c9","2671696d1afc4a478a60c27a4b2f13dd","bc9bbd6412e343bf8761ae96a814aac9","cca1861469384fa9a1220d0923650861","797b0fe55b5044d49641f0bbc17c2498","2ab55c29107c4d8b9653209d91983755","1121ac17c42140bcaed0970454800b4a","9c388a6c25ea47c49b2b22dd24b7e445","0e9b66113253434f9d23617b1f5d6119","dfa95837b9b445319294c3db64231abf","76c668899a984e38bffcd2f69c3ce14b","8a182f6ba34e4aa08d5275775f62b443","0710dbab770342a0815e25dac044eee3","1ccb953779ea497fbef3f5a5b3058a5c","d505968ea62246f39d9f0382f9aa0953","39389636f31b458186e2795290f4c42b","df962532bf794d88a5d3457cf311b6f5","850c24c1a9284b65917c756b72d60f91","15e39e3a50c747cfa45b618a3e7403b4","31a1528334bc49108cf33608e0512359","b87048f0c61e425591d559f5c214de93","b355333668b74860a2f81da3d8048af8","a03715fd2cda406b93acd98f5c6c41af","4a7266d0b24a4108868a6711256127c7","7e63a22f6c684f91bd656c3e65da6158","54f2446d49c346ce834de1befa18bd2c","4c1d62f9fc554466947c9973963495ff","081b6e87488d46c28c7462d66616ced6","8bbf6e08ec5d458fb20ee784300b2c87","e171e8b6a53f408fb0b8d5ae9b7609dc","649b37e2e92e40f4b4ac25dbbfad5f3f","65ec99f55e1b4fcb817e7e44b5a067b3","a7d0728325d04035b20696a66535cb46","752ee572888e40899bc7b836b79a5d58","841ef6d222b14448bf7b622a1c4da3b8","49fd4cd23f6e4a89986aca01a3392a73","4a4eadf90877428db3d75dd2edc0e466","68ec2ac98ca64edbb992163da859c3d8","0140b5c10c804ae4a654ff5b46139589","038709664d8741fca713c83e00b83966","1fa509840f8941bda369d75381130e26","483f08aa35a34726ad3225112935542a","5967ab3622fd49399135dd9f4d9a35b5","f46906bfd6a24951b44aa28a1a2b07b6","bdb7c09db6a04dce993e15167ee9c547","48a87645bef44cff906c6f3e801cea72","cd41166b07b64624af1ad5b0cad1f32a","0eff1c19e8e14e778a9fb52b48aabd01","76f3aee8199945408afe5bcad7159f15","c50428db770e4c5abde921a32badde1d","d7889da7c8404b6e8cd3f701eeb0c2b4","d09e05d7057e421caddca02aaaa0b666","314a15413f874f7a8aefa0dabb8e5508","ab54a4e50539487685869e29a24e7402","f2d47917e3bb4000bd09bd24543191a3","d8d489fd384c45db93a2fc2b3f12347b","7800df79b87b46a5af9c1489e690a177","41196335ecf148c5a8bc3c1a54dffb9b","afc72f9285bf4fee9fadb058f3201290","7e4672c8555849b98267fb4c3a491710","fd53ada9a1f84724b0384ffb7fe66732","a8bcf85fcb8f4da885d7ec433588ad4c","31349f8606db4827bdc9a7d87698f7a3","51c801bf4d1f40339e96b15e569411dd","eacffd8c09b448549f29d942ab0c9061","9d1f385e5dae46f59b78e748a7280a4f","7ff9020b41e544eab4f86c2e45a8d092","7492fb9555d44b5b9d847526ba284490","70b064e3df314bcdae4502e5a67a7c28","4cfee8f819624c95a3d74ba9558b64b7","478368ea1ccd43d8830d0f75fc67e2a1","0574c046687f4eada2317639551183b8","434dfeda872f4ae290d960c89104a486","9b4161a96f21490b8c858602f65844a6","bc0eff8590fa48719d3960164c641e79","157087b33015475a8226b4c86aed416d","229fb0b6bb0c4407b30a06641776df57","a48b779ebb864f3a8955ff279b2f44bd","db0a2dd0d4e94cfe91deaf41901ee8d3","7e2b25e3ff3e4c4da633e9b0de58922b","518fefd15e414aba908e29b6842a4e90","046639e8901e42c280a874f728c9d03a","1f2de106078446218506020442c7aeaf","54d6972637724629bf393072c086d813","f3329ed867054545b3a045af9c75de05","1ae970fd760d47978f3af86038f26f0f","7edc99ebc93e436aadcd7b813f78d92c","c811209d81e34863969e6f4f0a2c2877","a81608e6abd74f7387ee004a8e6977af","9607ab5565dd486f87241b1417b294c8","9614bf63dd2e4d4c9802a74f113d5c14","aa4b5dac1cbd4d0aa5ed74b367467c3d","3b43a8c3c26c4d2490237f7ac0aa9d61","0b590d19f70944e0960d539695a49c3f","f84dd6047ded477e8bdea1495560bc8c","cc9402b5f52341e58d09928c67e17d21","a91a15d761704f908840a9ac48e69543","b03df882d8424bbdbbb00857d82a79a2","f2073ca8c1c6467cad9ca2402d16c575","aaf89a5a64a54701819bb8e2e2984bb6","8dd773d1eefd469cae10f3bd3e9e627a","18d09157b6e947929301a0855e87e71c","d2c5844fae45496a9a6be73d7d1fd484","9d9bb1572c4746c7a294f2fc87bb9702","4255d387aac448ab8dad7a3d2b5e7a52","decc4490cb19434e8f62398b60bea037","00776136935f4d7e912c4ea7b18422a9","1f5cf831a673493ab7e75334826dd039","4045711b6e804a70bdf635c260c96f32","d6e0199799634196972c73eba67946f6","59bb7e0151914493b281a805ac80fda0","b583d6dbaa8f4ce9b8cdabc5ff9fd78a","56e8a5d918b74c02ae520a9d0e23457b","ddf6f88c0bef47058d474c326a90e53e","0fa75cf0dcae426ab3b4ccff7e8017ac"]},"id":"c001d0d1","outputId":"15558db7-1b2b-4729-ef77-28e7bc86c435","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:13:57.589167Z","iopub.execute_input":"2026-10-09T02:13:57.589924Z","iopub.status.idle":"2026-10-09T02:14:21.115571Z","shell.execute_reply.started":"2026-10-09T02:13:57.589900Z","shell.execute_reply":"2026-10-09T02:14:21.114592Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"Fetching 11 files:   0%|          | 0/11 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"be5ef52e5b7a4daa8308091fb8768737"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"byte_vocab.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"ce475120eef346fbbc46984cd136cb5e"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"README.md:   0%|          | 0.00/834 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"c3458b8c5e444f21bb8721ab7ec32806"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"special_tokens_map.json:   0%|          | 0.00/75.0 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"1961cb24bdaa47a4837fa643feb41a4d"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"coverage.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"73b56d00da79442c9ffb59cdc58253d0"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"latest.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"a7d48cc64a6f4a98a2619875871250e3"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"generation_config.json:   0%|          | 0.00/160 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"110009a91cbc4f7cb786e3e7f8569d83"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"model.safetensors:   0%|          | 0.00/361M [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"b16049196a284e0d86e534e21c0cb230"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"config.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"e72202305cbf4988ad5654e0d59dbf56"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"tokenization_byte.py: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"521f19db80c14664ac32e048b443156a"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"tokenizer_config.json:   0%|          | 0.00/932 [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"0177dccb619d47349508763252a28c6e"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"training_config.json: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"c1a6cfc127114df8af029619b67b3388"}},"metadata":{}},{"name":"stdout","text":"Pinned model snapshot: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model\nResolved Hub revision: 92c601560cc2179e3bfc33e067fdd3bf9254e163\nArchitecture: Qwen3ForCausalLM\nModel type: qwen3\nLayers: 24\nHidden size: 640\nMLP size: 1440\nAttention heads: 10\nKV heads: 2\nHead dim: 64\nVocab: 259\nNative context: 2048\nCheckpoint step: 900000\nLoading Void from pinned snapshot /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model ...\nLoaded: 90,148,352 parameters\nContext: 2048\nVocab: 259\nBOS/PAD/EOS: 257 256 258\n","output_type":"stream"}],"execution_count":10},{"id":"a5870e32","cell_type":"markdown","source":"## 5. Prompt inference","metadata":{"id":"a5870e32"}},{"id":"165cf16f","cell_type":"code","source":"@torch.inference_mode()\ndef generate_void(\n    prompt,\n    max_new_bytes=128,\n    temperature=0.0,\n    top_p=1.0,\n    repetition_penalty=1.15,\n    seed=42,\n):\n    model, tok = load_void()\n\n    bos_id = tok.bos_token_id if tok.bos_token_id is not None else 257\n    pad_id = tok.pad_token_id if tok.pad_token_id is not None else 256\n    eos_id = tok.eos_token_id if tok.eos_token_id is not None else 258\n\n    encoded = tok.encode(prompt, add_special_tokens=False)\n    max_context = int(getattr(model.config, \"max_position_embeddings\", 2048))\n\n    # Training records begin with BOS, so prepend it for free generation.\n    encoded = encoded[-(max_context - 1):]\n    input_ids = torch.tensor(\n        [[bos_id] + encoded],\n        dtype=torch.long,\n        device=DEVICE,\n    )\n\n    torch.manual_seed(seed)\n    if torch.cuda.is_available():\n        torch.cuda.manual_seed_all(seed)\n\n    do_sample = temperature is not None and float(temperature) > 0\n    gen_kwargs = dict(\n        max_new_tokens=int(max_new_bytes),\n        do_sample=do_sample,\n        use_cache=True,\n        pad_token_id=pad_id,\n        eos_token_id=eos_id,\n        repetition_penalty=float(repetition_penalty),\n    )\n    if do_sample:\n        gen_kwargs.update(\n            temperature=float(temperature),\n            top_p=float(top_p),\n        )\n\n    if torch.cuda.is_available():\n        amp_ctx = torch.autocast(\"cuda\", dtype=TORCH_DTYPE)\n    else:\n        import contextlib\n        amp_ctx = contextlib.nullcontext()\n\n    with amp_ctx:\n        out = model.generate(input_ids, **gen_kwargs)\n\n    new_ids = out[0, input_ids.shape[1]:].detach().cpu().tolist()\n    continuation = tok.decode(new_ids, skip_special_tokens=True)\n\n    return {\n        \"prompt\": prompt,\n        \"continuation\": continuation,\n        \"text\": prompt + continuation,\n        \"generated_bytes\": len(new_ids),\n    }\n\n\n# Small smoke test:\nresult = generate_void(\"How many r's are there in strawberry? Answer:\", max_new_bytes=32) # not there yet ;)\nprint(result[\"text\"])","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"165cf16f","outputId":"e5eb389a-2804-49da-ef75-5169a7883e47","trusted":true,"execution":{"iopub.status.busy":"2026-10-09T02:14:21.116668Z","iopub.execute_input":"2026-10-09T02:14:21.117299Z","iopub.status.idle":"2026-10-09T02:14:23.745349Z","shell.execute_reply.started":"2026-10-09T02:14:21.117272Z","shell.execute_reply":"2026-10-09T02:14:23.744520Z"},"jupyter":{"source_hidden":true}},"outputs":[{"name":"stdout","text":"How many r's are there in strawberry? Answer: There are 12 r's in strawberry.\n","output_type":"stream"}],"execution_count":11},{"id":"c5dbf0d2-1909-419d-974e-745df9f9a78a","cell_type":"code","source":"result = generate_void(\"a dataset is\", max_new_bytes=128, temperature=0.0)\nprint(result[\"text\"])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-09T02:14:23.746447Z","iopub.execute_input":"2026-10-09T02:14:23.746795Z","iopub.status.idle":"2026-10-09T02:14:28.162431Z","shell.execute_reply.started":"2026-10-09T02:14:23.746758Z","shell.execute_reply":"2026-10-09T02:14:28.161716Z"}},"outputs":[{"name":"stdout","text":"a dataset is a collection of data that contains information about a particular person, place, or event. It can be a list of names, addresses\n","output_type":"stream"}],"execution_count":12},{"id":"d2dadbc5","cell_type":"markdown","source":"### Optional interactive prompt UI","metadata":{"id":"d2dadbc5"}},{"id":"51221b42","cell_type":"code","source":"def launch_prompt_ui():\n    import ipywidgets as widgets\n    from IPython.display import display, clear_output\n\n    prompt_box = widgets.Textarea(\n        value=\"The capital of France is\",\n        description=\"Prompt:\",\n        layout=widgets.Layout(width=\"95%\", height=\"110px\"),\n    )\n    length_box = widgets.IntSlider(\n        value=128, min=8, max=512, step=8, description=\"New bytes:\"\n    )\n    temperature_box = widgets.FloatSlider(\n        value=0.0, min=0.0, max=1.5, step=0.05, description=\"Temp:\"\n    )\n    button = widgets.Button(description=\"Run Void\", button_style=\"primary\")\n    output = widgets.Output()\n\n    def _run(_):\n        with output:\n            clear_output(wait=True)\n            result = generate_void(\n                prompt_box.value,\n                max_new_bytes=length_box.value,\n                temperature=temperature_box.value,\n            )\n            print(result[\"text\"])\n\n    button.on_click(_run)\n    display(\n        prompt_box,\n        widgets.HBox([length_box, temperature_box]),\n        button,\n        output,\n    )\n\n# launch_prompt_ui()\n","metadata":{"id":"51221b42","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:14:28.163356Z","iopub.execute_input":"2026-10-09T02:14:28.163751Z","iopub.status.idle":"2026-10-09T02:14:28.170041Z","shell.execute_reply.started":"2026-10-09T02:14:28.163715Z","shell.execute_reply":"2026-10-09T02:14:28.169129Z"}},"outputs":[],"execution_count":13},{"id":"d6db57f9","cell_type":"markdown","source":"## 6. Benchmark infrastructure\n\nEach benchmark task runs in an isolated subprocess so a failed task or CUDA OOM does not poison the next model load.\n\nVoid has no recurrent compute modes, so there is one benchmark target and one set of results. The runner starts from a conservative fixed batch and automatically halves it on CUDA OOM for the lm-eval tasks.\n","metadata":{"id":"d6db57f9"}},{"id":"f597afc8","cell_type":"code","source":"from huggingface_hub import hf_hub_download\nfrom collections import deque\n\nBANANA_REPO = \"BananaMind/BananaMind-Base-Bench-1.1\"\nARITH3_REPO = \"AxiomicLabs/Arithmark-3.0\"\nBANANA_REVISION = \"d4aade51312889e8580963e1ce960c6eaef1a450\"\nARITH3_REVISION = \"main\"\n\nLM_EVAL_TASKS = [\"hellaswag\", \"piqa\", \"arc_easy\", \"arc_challenge\"]\n\nLM_EVAL_BATCH = 512\nBANANA_BATCH = 32\nARITH3_BATCH = 32\nLM_EVAL_MAX_LENGTH = BENCH_MAX_LENGTH\n\ndef benchmark_env():\n    if ACTIVE_MODEL is None:\n        raise RuntimeError(\"Pinned Void snapshot is not prepared. Run the model-loading cell first.\")\n    env = os.environ.copy()\n    env[\"HF_HUB_DISABLE_IMPLICIT_TOKEN\"] = \"1\"\n    if HF_TOKEN:\n        env[\"HF_TOKEN\"] = HF_TOKEN\n    else:\n        for key in (\"HF_TOKEN\", \"HUGGING_FACE_HUB_TOKEN\"):\n            env.pop(key, None)\n    env.setdefault(\"TOKENIZERS_PARALLELISM\", \"false\")\n    env.setdefault(\"PYTORCH_CUDA_ALLOC_CONF\", \"expandable_segments:True\")\n    return env\n\ndef newest_json(folder):\n    folder = Path(folder)\n    files = sorted(\n        folder.rglob(\"*.json\"),\n        key=lambda p: p.stat().st_mtime,\n        reverse=True,\n    )\n    return files[0] if files else None\n\ndef flatten_dict(obj, prefix=\"\"):\n    out = {}\n    if isinstance(obj, dict):\n        for k, v in obj.items():\n            name = f\"{prefix}.{k}\" if prefix else str(k)\n            if isinstance(v, (dict, list)):\n                out.update(flatten_dict(v, name))\n            else:\n                out[name] = v\n    elif isinstance(obj, list):\n        for i, v in enumerate(obj):\n            name = f\"{prefix}[{i}]\"\n            if isinstance(v, (dict, list)):\n                out.update(flatten_dict(v, name))\n            else:\n                out[name] = v\n    return out\n\ndef print_json_path(path):\n    if path is None:\n        print(\"No JSON result file found.\")\n        return None\n    print(\"Result JSON:\", path)\n    with open(path, \"r\", encoding=\"utf-8\") as f:\n        return json.load(f)\n\ndef free_prompt_model_before_benchmark():\n    if \"void_model\" in globals() and globals().get(\"void_model\") is not None:\n        print(\"Freeing interactive Void model before benchmark subprocess...\")\n        unload_void()\n    gc.collect()\n    if torch.cuda.is_available():\n        torch.cuda.empty_cache()\n\ndef _run_logged(cmd, *, cwd=None, check=True):\n    env = benchmark_env()\n    print(\"\\nCOMMAND:\", \" \".join(map(str, cmd)), flush=True)\n    proc = subprocess.Popen(\n        [str(x) for x in cmd],\n        env=env,\n        cwd=cwd,\n        stdout=subprocess.PIPE,\n        stderr=subprocess.STDOUT,\n        text=True,\n        bufsize=1,\n    )\n    tail = deque(maxlen=160)\n    assert proc.stdout is not None\n    for line in proc.stdout:\n        print(line, end=\"\", flush=True)\n        tail.append(line.rstrip(\"\\n\"))\n\n    rc = proc.wait()\n    result = {\n        \"returncode\": rc,\n        \"tail\": \"\\n\".join(tail),\n        \"command\": [str(x) for x in cmd],\n    }\n    if check and rc != 0:\n        raise RuntimeError(\n            f\"Benchmark child exited with code {rc}.\\n\\n\"\n            f\"Last child-process lines:\\n{result['tail']}\"\n        )\n    return result\n\ndef _run(cmd, *, cwd=None):\n    return _run_logged(cmd, cwd=cwd, check=True)\n\nprint(\"lm-eval starting batch:\", LM_EVAL_BATCH)\nprint(\"BananaMind batch:\", BANANA_BATCH)\nprint(\"ArithMark batch:\", ARITH3_BATCH)\n","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"f597afc8","outputId":"83e5478f-f24e-4636-cb27-9e8150b52b2d","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:14:28.171033Z","iopub.execute_input":"2026-10-09T02:14:28.171322Z","iopub.status.idle":"2026-10-09T02:14:28.189682Z","shell.execute_reply.started":"2026-10-09T02:14:28.171290Z","shell.execute_reply":"2026-10-09T02:14:28.188792Z"}},"outputs":[{"name":"stdout","text":"lm-eval starting batch: 512\nBananaMind batch: 32\nArithMark batch: 32\n","output_type":"stream"}],"execution_count":14},{"id":"8e82b5be","cell_type":"markdown","source":"### Full-benchmark fix for Void's byte tokenizer\n\nThe same byte-level issue from the Fig benchmark can affect Void: a short natural-language answer may become more than 1,024 byte IDs.\n\nThe adapter below preserves stock `lm-eval` behavior for normal requests. Only a continuation longer than the configured benchmark window uses the fallback:\n\n- split the continuation into disjoint scored segments;\n- retain as much immediately preceding context as fits;\n- score every continuation byte exactly once;\n- sum segment log-likelihoods;\n- AND greedy correctness across segments;\n- record fallback counts in the result JSON.\n\nThis keeps PIQA and other tasks from crashing on an oversized byte continuation without silently truncating the answer.\n","metadata":{"id":"8e82b5be"}},{"id":"c30807ef","cell_type":"markdown","source":"## 7. Open SLM standard benchmarks — HellaSwag, PIQA, ARC Easy, ARC Challenge","metadata":{"id":"c30807ef"}},{"id":"4c904386","cell_type":"code","source":"LM_DRIVER = ROOT / \"_run_void_lm_eval.py\"\n\nLM_DRIVER.write_text(r\"\"\"\nimport argparse, json, traceback, sys\nimport lm_eval\nfrom lm_eval.models.huggingface import HFLM\n\np = argparse.ArgumentParser()\np.add_argument(\"--model\", required=True)\np.add_argument(\"--revision\", default=\"main\")\np.add_argument(\"--task\", required=True)\np.add_argument(\"--device\", default=\"cuda:0\")\np.add_argument(\"--dtype\", default=\"bfloat16\")\np.add_argument(\"--batch-size\", default=\"16\")\np.add_argument(\"--max-length\", type=int, default=1024)\np.add_argument(\"--long-cont-stride\", type=int, default=256)\np.add_argument(\"--limit\", type=float, default=None)\np.add_argument(\"--out\", required=True)\na = p.parse_args()\n\n\nclass VoidHFLM(HFLM):\n    '''\n    HFLM adapter for Void's byte tokenizer.\n\n    Stock behavior is unchanged when the continuation fits in max_length.\n    Only oversized byte continuations are scored in disjoint sliding segments.\n    '''\n\n    def __init__(self, *args, long_cont_stride=256, **kwargs):\n        super().__init__(*args, **kwargs)\n        stride = int(long_cont_stride)\n        if stride < 1:\n            raise ValueError(\"long_cont_stride must be >= 1\")\n        self.long_cont_stride = min(stride, max(1, self.max_length - 1))\n        self.void_long_request_count = 0\n        self.void_long_continuation_tokens = 0\n        self.void_max_continuation_tokens = 0\n\n    def _score_long_request(self, request):\n        request_str, context_enc, continuation_enc = request\n\n        if not context_enc:\n            raise ValueError(\"Void long-continuation scoring requires non-empty context\")\n        if not continuation_enc:\n            raise ValueError(\"Void long-continuation scoring requires non-empty continuation\")\n\n        self.void_long_request_count += 1\n        self.void_long_continuation_tokens += len(continuation_enc)\n        self.void_max_continuation_tokens = max(\n            self.void_max_continuation_tokens,\n            len(continuation_enc),\n        )\n\n        total_logprob = 0.0\n        all_greedy = True\n        offset = 0\n\n        while offset < len(continuation_enc):\n            segment = continuation_enc[offset : offset + self.long_cont_stride]\n\n            # [history | segment] -> remove final token for causal inputs.\n            # Keep exactly enough history so model inputs remain <= max_length.\n            prefix = context_enc + continuation_enc[:offset]\n            context_room = max(1, self.max_length + 1 - len(segment))\n            context_window = prefix[-context_room:]\n\n            answer = super()._loglikelihood_tokens(\n                [(None, context_window, segment)],\n                disable_tqdm=True,\n                override_bs=1,\n            )[0]\n\n            total_logprob += float(answer[0])\n            all_greedy = all_greedy and bool(answer[1])\n            offset += len(segment)\n\n        final = (total_logprob, all_greedy)\n\n        if request_str is not None:\n            self.cache_hook.add_partial(\"loglikelihood\", request_str, final)\n\n        return final\n\n    def _loglikelihood_tokens(\n        self,\n        requests,\n        disable_tqdm=False,\n        override_bs=None,\n    ):\n        outputs = [None] * len(requests)\n        short_requests = []\n        short_indices = []\n        long_items = []\n\n        for i, req in enumerate(requests):\n            continuation_enc = req[2]\n            if len(continuation_enc) <= self.max_length:\n                short_indices.append(i)\n                short_requests.append(req)\n            else:\n                long_items.append((i, req))\n\n        if short_requests:\n            short_outputs = super()._loglikelihood_tokens(\n                short_requests,\n                disable_tqdm=disable_tqdm,\n                override_bs=override_bs,\n            )\n            for i, ans in zip(short_indices, short_outputs):\n                outputs[i] = ans\n\n        if long_items:\n            print(\n                f\"Void byte-level fallback: {len(long_items)} request(s) have \"\n                f\"continuations longer than {self.max_length} IDs; \"\n                f\"sliding stride={self.long_cont_stride}.\",\n                flush=True,\n            )\n\n        for i, req in long_items:\n            outputs[i] = self._score_long_request(req)\n\n        assert all(x is not None for x in outputs)\n        return outputs\n\n\nprint(\n    \"task =\", a.task,\n    \"batch =\", a.batch_size,\n    \"max_length =\", a.max_length,\n    \"long_cont_stride =\", a.long_cont_stride,\n    flush=True,\n)\n\ntry:\n    lm = VoidHFLM(\n        pretrained=a.model,\n        revision=a.revision,\n        device=a.device,\n        dtype=a.dtype,\n        batch_size=int(a.batch_size),\n        max_length=int(a.max_length),\n        truncation=True,\n        trust_remote_code=True,\n        use_fast_tokenizer=False,\n        long_cont_stride=int(a.long_cont_stride),\n    )\n\n    result = lm_eval.simple_evaluate(\n        model=lm,\n        tasks=[a.task],\n        num_fewshot=0,\n        limit=a.limit,\n        log_samples=False,\n    )\n\n    payload = {\n        \"results\": result.get(\"results\", {}),\n        \"versions\": result.get(\"versions\", {}),\n        \"n-shot\": result.get(\"n-shot\", {}),\n        \"n-samples\": result.get(\"n-samples\", {}),\n        \"void_eval\": {\n            \"max_length\": int(a.max_length),\n            \"long_continuation_policy\": \"sliding_disjoint_segments\",\n            \"long_continuation_stride\": int(lm.long_cont_stride),\n            \"long_request_count\": int(lm.void_long_request_count),\n            \"long_continuation_tokens\": int(lm.void_long_continuation_tokens),\n            \"max_continuation_tokens\": int(lm.void_max_continuation_tokens),\n        },\n    }\n\n    with open(a.out, \"w\", encoding=\"utf-8\") as f:\n        json.dump(payload, f, indent=2, default=str)\n\n    print(json.dumps(payload[\"results\"], indent=2, default=str), flush=True)\n    print(\n        \"Void eval metadata:\",\n        json.dumps(payload[\"void_eval\"], indent=2),\n        flush=True,\n    )\n\nexcept BaseException:\n    traceback.print_exc()\n    sys.stdout.flush()\n    sys.stderr.flush()\n    raise\n\"\"\", encoding=\"utf-8\")\n\n\ndef _metric_value(\n    task_result,\n    preferred=(\"acc_norm,none\", \"acc_norm\", \"acc,none\", \"acc\"),\n):\n    for key in preferred:\n        if key in task_result and isinstance(task_result[key], (int, float)):\n            value = float(task_result[key])\n            return value * 100 if abs(value) <= 1.0 else value\n    return None\n\n\ndef _looks_like_oom(text):\n    t = (text or \"\").lower()\n    markers = (\n        \"out of memory\",\n        \"cuda oom\",\n        \"cuda error: out of memory\",\n        \"cublas_status_alloc_failed\",\n        \"failed to allocate\",\n    )\n    return any(m in t for m in markers)\n\n\ndef _batch_retry_sequence(start):\n    start = max(1, int(start))\n    values = []\n    x = start\n    while x >= 1:\n        if x not in values:\n            values.append(x)\n        if x == 1:\n            break\n        x = max(1, x // 2)\n    return values\n\n\ndef run_one_open_slm_task(\n    task,\n    *,\n    limit=None,\n    batch_size=None,\n    long_cont_stride=256,\n):\n    out_dir = ROOT / \"open_slm\" / task\n    out_dir.mkdir(parents=True, exist_ok=True)\n    result_file = out_dir / \"results.json\"\n    result_file.unlink(missing_ok=True)\n\n    start_batch = LM_EVAL_BATCH if batch_size is None else int(batch_size)\n\n    for attempt_batch in _batch_retry_sequence(start_batch):\n        cmd = [\n            sys.executable,\n            str(LM_DRIVER),\n            \"--model\", ACTIVE_MODEL,\n            \"--revision\", RESOLVED_MODEL_REVISION,\n            \"--task\", task,\n            \"--device\", DEVICE,\n            \"--dtype\", BENCH_DTYPE,\n            \"--batch-size\", str(attempt_batch),\n            \"--max-length\", str(LM_EVAL_MAX_LENGTH),\n            \"--long-cont-stride\", str(int(long_cont_stride)),\n            \"--out\", str(result_file),\n        ]\n        if limit is not None:\n            cmd += [\"--limit\", str(limit)]\n\n        print(f\"\\n[{task}] trying fixed batch={attempt_batch}\")\n        proc = _run_logged(cmd, check=False)\n\n        if proc[\"returncode\"] == 0:\n            if not result_file.exists():\n                raise RuntimeError(\n                    f\"{task} exited successfully but did not create {result_file}\"\n                )\n            payload = json.loads(result_file.read_text())\n            return {\n                \"task\": task,\n                \"batch_size\": attempt_batch,\n                \"raw\": payload,\n                \"path\": str(result_file),\n                \"void_eval\": payload.get(\"void_eval\", {}),\n            }\n\n        if _looks_like_oom(proc[\"tail\"]) and attempt_batch > 1:\n            print(\n                f\"\\n[{task}] CUDA OOM at batch={attempt_batch}; \"\n                \"retrying smaller batch.\"\n            )\n            gc.collect()\n            if torch.cuda.is_available():\n                torch.cuda.empty_cache()\n            continue\n\n        raise RuntimeError(\n            f\"{task} failed at batch={attempt_batch} \"\n            f\"with exit code {proc['returncode']}.\\n\\n\"\n            f\"Real child traceback/output:\\n{proc['tail']}\"\n        )\n\n    raise RuntimeError(f\"{task} failed even at batch size 1.\")\n\n\ndef run_open_slm(\n    tasks=LM_EVAL_TASKS,\n    limit=None,\n    batch_size=None,\n    long_cont_stride=256,\n):\n    \"\"\"\n    Void standard benchmark runner.\n\n    * one task per subprocess\n    * 0-shot lm-eval\n    * fixed batch with CUDA-OOM backoff\n    * byte-level long-continuation fallback\n    * same default 1024 evaluation window as the Fig benchmark\n    \"\"\"\n    free_prompt_model_before_benchmark()\n\n    summary = {\n        \"model\": \"Void\",\n        \"model_id\": MODEL_ID,\n        \"revision\": RESOLVED_MODEL_REVISION or MODEL_REVISION,\n    }\n    raw_by_task = {}\n    paths = {}\n    batches = {}\n    eval_metadata = {}\n\n    for task in tasks:\n        print(\"\\n\" + \"=\" * 80)\n        print(f\"VOID — {task}\")\n        print(\"=\" * 80)\n\n        item = run_one_open_slm_task(\n            task,\n            limit=limit,\n            batch_size=batch_size,\n            long_cont_stride=long_cont_stride,\n        )\n        payload = item[\"raw\"]\n        raw_by_task[task] = payload\n        paths[task] = item[\"path\"]\n        batches[task] = item[\"batch_size\"]\n        eval_metadata[task] = item.get(\"void_eval\", {})\n\n        r = payload.get(\"results\", {}).get(task, {})\n        summary[task] = _metric_value(r)\n\n        progress_dir = ROOT / \"open_slm\"\n        (progress_dir / \"summary_partial.json\").write_text(\n            json.dumps(\n                {\n                    \"summary\": summary,\n                    \"batches\": batches,\n                    \"paths\": paths,\n                    \"void_eval\": eval_metadata,\n                },\n                indent=2,\n            ),\n            encoding=\"utf-8\",\n        )\n\n        display(pd.DataFrame([summary]))\n\n    combined_path = ROOT / \"open_slm\" / \"summary.json\"\n    combined_path.write_text(\n        json.dumps(\n            {\n                \"summary\": summary,\n                \"batches\": batches,\n                \"paths\": paths,\n                \"void_eval\": eval_metadata,\n            },\n            indent=2,\n        ),\n        encoding=\"utf-8\",\n    )\n\n    print(\"\\nCompleted all requested Open SLM tasks.\")\n    display(pd.DataFrame([summary]))\n    print(\"Combined summary:\", combined_path)\n\n    return {\n        \"summary\": summary,\n        \"raw\": raw_by_task,\n        \"paths\": paths,\n        \"batches\": batches,\n        \"void_eval\": eval_metadata,\n        \"path\": str(combined_path),\n    }\n\n\n# Recommended smoke test:\n# piqa_smoke = run_open_slm(tasks=[\"piqa\"], limit=50)\n#\n# Full standard suite:\n# open_scores = run_open_slm()\n","metadata":{"id":"4c904386","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:14:28.190904Z","iopub.execute_input":"2026-10-09T02:14:28.191328Z","iopub.status.idle":"2026-10-09T02:14:28.213165Z","shell.execute_reply.started":"2026-10-09T02:14:28.191226Z","shell.execute_reply":"2026-10-09T02:14:28.212366Z"}},"outputs":[],"execution_count":15},{"id":"90eec483","cell_type":"code","source":"# Pure-Python sanity check for the long-continuation window geometry.\n# This does not load the model.\n\ndef _void_segment_geometry(\n    context_len,\n    continuation_len,\n    max_length=BENCH_MAX_LENGTH,\n    stride=256,\n):\n    offset = 0\n    rows = []\n\n    while offset < continuation_len:\n        seg_len = min(stride, continuation_len - offset)\n        prefix_len = context_len + offset\n        context_room = max(1, max_length + 1 - seg_len)\n        used_context = min(prefix_len, context_room)\n        model_input_len = used_context + seg_len - 1\n\n        rows.append({\n            \"offset\": offset,\n            \"segment_len\": seg_len,\n            \"used_context\": used_context,\n            \"model_input_len\": model_input_len,\n        })\n\n        assert seg_len <= max_length\n        assert 1 <= used_context\n        assert model_input_len <= max_length\n        offset += seg_len\n\n    assert sum(r[\"segment_len\"] for r in rows) == continuation_len\n    return rows\n\n\n# Reproduce the 1,106-byte PIQA case that motivated the original Fig fix.\n_geometry = _void_segment_geometry(\n    context_len=128,\n    continuation_len=1106,\n    max_length=BENCH_MAX_LENGTH,\n    stride=256,\n)\ndisplay(pd.DataFrame(_geometry))\nprint(\n    \"PASS: all 1,106 continuation bytes are covered exactly once and every \"\n    f\"model input stays within {BENCH_MAX_LENGTH:,} IDs.\"\n)\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":244},"id":"90eec483","outputId":"852e0838-9b26-42ea-99e2-218eff9e8651","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:14:28.215249Z","iopub.execute_input":"2026-10-09T02:14:28.215564Z","iopub.status.idle":"2026-10-09T02:14:28.287572Z","shell.execute_reply.started":"2026-10-09T02:14:28.215527Z","shell.execute_reply":"2026-10-09T02:14:28.286815Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"   offset  segment_len  used_context  model_input_len\n0       0          256           128              383\n1     256          256           384              639\n2     512          256           640              895\n3     768          256           896             1151\n4    1024           82          1152             1233","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>offset</th>\n      <th>segment_len</th>\n      <th>used_context</th>\n      <th>model_input_len</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>256</td>\n      <td>128</td>\n      <td>383</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>256</td>\n      <td>256</td>\n      <td>384</td>\n      <td>639</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>512</td>\n      <td>256</td>\n      <td>640</td>\n      <td>895</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>768</td>\n      <td>256</td>\n      <td>896</td>\n      <td>1151</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1024</td>\n      <td>82</td>\n      <td>1152</td>\n      <td>1233</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"PASS: all 1,106 continuation bytes are covered exactly once and every model input stays within 2,048 IDs.\n","output_type":"stream"}],"execution_count":16},{"id":"b0d72837","cell_type":"markdown","source":"## 8. BananaMind Base Bench 1.1 — official runner\n\nThis downloads `benchmark.py` directly from the benchmark dataset repository and executes it unchanged. The helper checks the runner's `--help` output first and only passes flags that version supports.","metadata":{"id":"b0d72837"}},{"id":"f24e0ce8","cell_type":"code","source":"def _add_flag_if_supported(cmd, help_text, flag, value=None):\n    if flag in help_text:\n        cmd.append(flag)\n        if value is not None:\n            cmd.append(str(value))\n\ndef run_bananamind(batch_size=BANANA_BATCH):\n    out_dir = ROOT / \"bananamind_base_1_1\"\n    out_dir.mkdir(parents=True, exist_ok=True)\n\n    script = Path(hf_hub_download(\n        repo_id=BANANA_REPO,\n        repo_type=\"dataset\",\n        filename=\"benchmark.py\",\n        revision=BANANA_REVISION,\n        token=HF_TOKEN,\n    ))\n\n    help_proc = subprocess.run(\n        [sys.executable, str(script), \"--help\"],\n        env=benchmark_env(),\n        capture_output=True,\n        text=True,\n        check=True,\n    )\n    help_text = help_proc.stdout + \"\\n\" + help_proc.stderr\n\n    cmd = [sys.executable, str(script)]\n    _add_flag_if_supported(cmd, help_text, \"--model\", ACTIVE_MODEL)\n    _add_flag_if_supported(cmd, help_text, \"--model-revision\", RESOLVED_MODEL_REVISION)\n    _add_flag_if_supported(cmd, help_text, \"--dataset-revision\", BANANA_REVISION)\n    _add_flag_if_supported(\n        cmd, help_text, \"--device\",\n        \"cuda\" if torch.cuda.is_available() else \"cpu\"\n    )\n    _add_flag_if_supported(cmd, help_text, \"--dtype\", BENCH_DTYPE)\n    _add_flag_if_supported(cmd, help_text, \"--batch-size\", batch_size)\n    _add_flag_if_supported(cmd, help_text, \"--threads\", 2)\n    _add_flag_if_supported(cmd, help_text, \"--out-dir\", str(out_dir))\n\n    # If a runner exposes a context limit, match the standard suite.\n    _add_flag_if_supported(cmd, help_text, \"--max-context\", BENCH_MAX_LENGTH)\n    _add_flag_if_supported(cmd, help_text, \"--max-length\", BENCH_MAX_LENGTH)\n\n    if \"--trust-remote-code\" in help_text and \"--no-trust-remote-code\" not in help_text:\n        cmd.append(\"--trust-remote-code\")\n\n    _run(cmd, cwd=out_dir)\n\n    path = newest_json(out_dir)\n    payload = print_json_path(path)\n\n    if payload is not None:\n        flat = flatten_dict(payload)\n        interesting = {\n            k: v for k, v in flat.items()\n            if any(x in k.lower() for x in (\"elo\", \"accuracy\", \"weighted\", \"overall\"))\n            and isinstance(v, (int, float))\n        }\n        if interesting:\n            display(pd.DataFrame([interesting]))\n\n    return {\n        \"raw\": payload,\n        \"path\": str(path) if path else None,\n        \"output_dir\": str(out_dir),\n    }\n\nbanana_scores = run_bananamind()\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["454a64c1b29c416fbd7be768488027b3","19f35d68a8aa4c679ff9bbcc8d145f0e","20eebefa4a5f4c1782e41d20a5199b76","f5a146767d5446ec9a62888d02c5d3df","bff1824dfcb1442d83c30440b39764dd","b575e4632508471398023e563cf1a5d0","2953c3a16a634ea6a22fe78ef706ab57","167e7db8138c4dca9037528709524a49","5f4d2371084748f8aa1e988fec6c09f6","8fcd4fcf166641deaf192294feed096d","10dcccd3ba18474cb6947da57da155d0"]},"id":"f24e0ce8","outputId":"d3cb65c3-9646-4186-9731-abb7b25e6a06","trusted":true,"jupyter":{"source_hidden":true},"scrolled":true,"execution":{"iopub.status.busy":"2026-10-09T02:14:28.288417Z","iopub.execute_input":"2026-10-09T02:14:28.288729Z","iopub.status.idle":"2026-10-09T02:14:57.822035Z","shell.execute_reply.started":"2026-10-09T02:14:28.288698Z","shell.execute_reply":"2026-10-09T02:14:57.821389Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"benchmark.py: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"304dc78ec2404d679bb0bcc02f0b327a"}},"metadata":{}},{"name":"stdout","text":"\nCOMMAND: /usr/bin/python3 /root/.cache/huggingface/hub/datasets--BananaMind--BananaMind-Base-Bench-1.1/snapshots/d4aade51312889e8580963e1ce960c6eaef1a450/benchmark.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --model-revision 92c601560cc2179e3bfc33e067fdd3bf9254e163 --dataset-revision d4aade51312889e8580963e1ce960c6eaef1a450 --device cuda --dtype float16 --batch-size 32 --threads 2 --out-dir /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1 --max-context 2048\nChecking --hf-token/HF_TOKEN access to BananaMind/BananaMind-Base-Bench-1.1...\nDataset access confirmed.\nLoading /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model on cuda as float16...\nScoring raw continuations with context length 2048; add_bos=False.\n[001/350] PASS language_completion-001 (pred=2, label=2)\n[002/350] PASS language_completion-002 (pred=1, label=1)\n[003/350] PASS language_completion-003 (pred=2, label=2)\n[004/350] PASS language_completion-004 (pred=3, label=3)\n[005/350] PASS language_completion-005 (pred=0, label=0)\n[006/350] PASS language_completion-006 (pred=1, label=1)\n[007/350] PASS language_completion-007 (pred=2, label=2)\n[008/350] PASS language_completion-008 (pred=3, label=3)\n[009/350] PASS language_completion-009 (pred=0, label=0)\n[010/350] PASS language_completion-010 (pred=1, label=1)\n[011/350] PASS language_completion-011 (pred=2, label=2)\n[012/350] PASS language_completion-012 (pred=3, label=3)\n[013/350] PASS language_completion-013 (pred=0, label=0)\n[014/350] PASS language_completion-014 (pred=1, label=1)\n[015/350] PASS language_completion-015 (pred=2, label=2)\n[016/350] FAIL language_completion-016 (pred=0, label=3)\n[017/350] PASS language_completion-017 (pred=0, label=0)\n[018/350] FAIL language_completion-018 (pred=0, label=1)\n[019/350] PASS language_completion-019 (pred=2, label=2)\n[020/350] PASS language_completion-020 (pred=3, label=3)\n[021/350] PASS language_completion-021 (pred=0, label=0)\n[022/350] PASS language_completion-022 (pred=1, label=1)\n[023/350] PASS language_completion-023 (pred=2, label=2)\n[024/350] PASS language_completion-024 (pred=3, label=3)\n[025/350] FAIL language_completion-025 (pred=3, label=0)\n[026/350] PASS language_completion-026 (pred=1, label=1)\n[027/350] PASS language_completion-027 (pred=2, label=2)\n[028/350] PASS language_completion-028 (pred=3, label=3)\n[029/350] PASS language_completion-029 (pred=0, label=0)\n[030/350] PASS language_completion-030 (pred=1, label=1)\n[031/350] PASS language_completion-031 (pred=2, label=2)\n[032/350] FAIL language_completion-032 (pred=0, label=3)\n[033/350] PASS language_completion-033 (pred=0, label=0)\n[034/350] PASS language_completion-034 (pred=1, label=1)\n[035/350] PASS language_completion-035 (pred=2, label=2)\n[036/350] PASS language_completion-036 (pred=3, label=3)\n[037/350] PASS language_completion-037 (pred=0, label=0)\n[038/350] PASS language_completion-038 (pred=1, label=1)\n[039/350] PASS language_completion-039 (pred=2, label=2)\n[040/350] PASS language_completion-040 (pred=3, label=3)\n[041/350] PASS language_completion-041 (pred=0, label=0)\n[042/350] FAIL language_completion-042 (pred=2, label=1)\n[043/350] PASS language_completion-043 (pred=2, label=2)\n[044/350] PASS language_completion-044 (pred=3, label=3)\n[045/350] PASS language_completion-045 (pred=0, label=0)\n[046/350] PASS language_completion-046 (pred=1, label=1)\n[047/350] PASS language_completion-047 (pred=2, label=2)\n[048/350] PASS language_completion-048 (pred=3, label=3)\n[049/350] PASS language_completion-049 (pred=0, label=0)\n[050/350] PASS language_completion-050 (pred=1, label=1)\n[051/350] PASS commonsense-001 (pred=2, label=2)\n[052/350] PASS commonsense-002 (pred=1, label=1)\n[053/350] PASS commonsense-003 (pred=2, label=2)\n[054/350] PASS commonsense-004 (pred=3, label=3)\n[055/350] PASS commonsense-005 (pred=0, label=0)\n[056/350] PASS commonsense-006 (pred=1, label=1)\n[057/350] PASS commonsense-007 (pred=2, label=2)\n[058/350] FAIL commonsense-008 (pred=0, label=3)\n[059/350] PASS commonsense-009 (pred=0, label=0)\n[060/350] PASS commonsense-010 (pred=1, label=1)\n[061/350] PASS commonsense-011 (pred=2, label=2)\n[062/350] PASS commonsense-012 (pred=3, label=3)\n[063/350] PASS commonsense-013 (pred=0, label=0)\n[064/350] PASS commonsense-014 (pred=1, label=1)\n[065/350] PASS commonsense-015 (pred=2, label=2)\n[066/350] PASS commonsense-016 (pred=3, label=3)\n[067/350] PASS commonsense-017 (pred=0, label=0)\n[068/350] PASS commonsense-018 (pred=1, label=1)\n[069/350] PASS commonsense-019 (pred=2, label=2)\n[070/350] PASS commonsense-020 (pred=3, label=3)\n[071/350] PASS commonsense-021 (pred=0, label=0)\n[072/350] PASS commonsense-022 (pred=1, label=1)\n[073/350] FAIL commonsense-023 (pred=3, label=2)\n[074/350] FAIL commonsense-024 (pred=0, label=3)\n[075/350] PASS commonsense-025 (pred=0, label=0)\n[076/350] FAIL commonsense-026 (pred=2, label=1)\n[077/350] PASS commonsense-027 (pred=2, label=2)\n[078/350] PASS commonsense-028 (pred=3, label=3)\n[079/350] FAIL commonsense-029 (pred=3, label=0)\n[080/350] PASS commonsense-030 (pred=1, label=1)\n[081/350] PASS commonsense-031 (pred=2, label=2)\n[082/350] FAIL commonsense-032 (pred=0, label=3)\n[083/350] FAIL commonsense-033 (pred=1, label=0)\n[084/350] PASS commonsense-034 (pred=1, label=1)\n[085/350] FAIL commonsense-035 (pred=3, label=2)\n[086/350] PASS commonsense-036 (pred=3, label=3)\n[087/350] PASS commonsense-037 (pred=0, label=0)\n[088/350] PASS commonsense-038 (pred=1, label=1)\n[089/350] FAIL commonsense-039 (pred=0, label=2)\n[090/350] PASS commonsense-040 (pred=3, label=3)\n[091/350] PASS commonsense-041 (pred=0, label=0)\n[092/350] FAIL commonsense-042 (pred=0, label=1)\n[093/350] PASS commonsense-043 (pred=2, label=2)\n[094/350] PASS commonsense-044 (pred=3, label=3)\n[095/350] PASS commonsense-045 (pred=0, label=0)\n[096/350] FAIL commonsense-046 (pred=0, label=1)\n[097/350] PASS commonsense-047 (pred=2, label=2)\n[098/350] PASS commonsense-048 (pred=3, label=3)\n[099/350] FAIL commonsense-049 (pred=2, label=0)\n[100/350] PASS commonsense-050 (pred=1, label=1)\n[101/350] PASS world_knowledge-001 (pred=2, label=2)\n[102/350] PASS world_knowledge-002 (pred=1, label=1)\n[103/350] FAIL world_knowledge-003 (pred=1, label=2)\n[104/350] PASS world_knowledge-004 (pred=3, label=3)\n[105/350] PASS world_knowledge-005 (pred=0, label=0)\n[106/350] PASS world_knowledge-006 (pred=1, label=1)\n[107/350] PASS world_knowledge-007 (pred=2, label=2)\n[108/350] PASS world_knowledge-008 (pred=3, label=3)\n[109/350] PASS world_knowledge-009 (pred=0, label=0)\n[110/350] PASS world_knowledge-010 (pred=1, label=1)\n[111/350] FAIL world_knowledge-011 (pred=0, label=2)\n[112/350] FAIL world_knowledge-012 (pred=2, label=3)\n[113/350] PASS world_knowledge-013 (pred=0, label=0)\n[114/350] PASS world_knowledge-014 (pred=1, label=1)\n[115/350] PASS world_knowledge-015 (pred=2, label=2)\n[116/350] PASS world_knowledge-016 (pred=3, label=3)\n[117/350] PASS world_knowledge-017 (pred=0, label=0)\n[118/350] PASS world_knowledge-018 (pred=1, label=1)\n[119/350] FAIL world_knowledge-019 (pred=1, label=2)\n[120/350] PASS world_knowledge-020 (pred=3, label=3)\n[121/350] PASS world_knowledge-021 (pred=0, label=0)\n[122/350] PASS world_knowledge-022 (pred=1, label=1)\n[123/350] PASS world_knowledge-023 (pred=2, label=2)\n[124/350] PASS world_knowledge-024 (pred=3, label=3)\n[125/350] PASS world_knowledge-025 (pred=0, label=0)\n[126/350] PASS world_knowledge-026 (pred=1, label=1)\n[127/350] FAIL world_knowledge-027 (pred=1, label=2)\n[128/350] PASS world_knowledge-028 (pred=3, label=3)\n[129/350] PASS world_knowledge-029 (pred=0, label=0)\n[130/350] PASS world_knowledge-030 (pred=1, label=1)\n[131/350] PASS world_knowledge-031 (pred=2, label=2)\n[132/350] FAIL world_knowledge-032 (pred=1, label=3)\n[133/350] PASS world_knowledge-033 (pred=0, label=0)\n[134/350] PASS world_knowledge-034 (pred=1, label=1)\n[135/350] PASS world_knowledge-035 (pred=2, label=2)\n[136/350] PASS world_knowledge-036 (pred=3, label=3)\n[137/350] PASS world_knowledge-037 (pred=0, label=0)\n[138/350] PASS world_knowledge-038 (pred=1, label=1)\n[139/350] FAIL world_knowledge-039 (pred=1, label=2)\n[140/350] PASS world_knowledge-040 (pred=3, label=3)\n[141/350] FAIL world_knowledge-041 (pred=2, label=0)\n[142/350] FAIL world_knowledge-042 (pred=0, label=1)\n[143/350] PASS world_knowledge-043 (pred=2, label=2)\n[144/350] PASS world_knowledge-044 (pred=3, label=3)\n[145/350] FAIL world_knowledge-045 (pred=1, label=0)\n[146/350] FAIL world_knowledge-046 (pred=0, label=1)\n[147/350] PASS world_knowledge-047 (pred=2, label=2)\n[148/350] FAIL world_knowledge-048 (pred=0, label=3)\n[149/350] PASS world_knowledge-049 (pred=0, label=0)\n[150/350] FAIL world_knowledge-050 (pred=0, label=1)\n[151/350] PASS context_tracking-001 (pred=0, label=0)\n[152/350] FAIL context_tracking-002 (pred=0, label=3)\n[153/350] PASS context_tracking-003 (pred=2, label=2)\n[154/350] PASS context_tracking-004 (pred=3, label=3)\n[155/350] FAIL context_tracking-005 (pred=1, label=0)\n[156/350] FAIL context_tracking-006 (pred=3, label=1)\n[157/350] PASS context_tracking-007 (pred=2, label=2)\n[158/350] FAIL context_tracking-008 (pred=1, label=3)\n[159/350] PASS context_tracking-009 (pred=0, label=0)\n[160/350] FAIL context_tracking-010 (pred=2, label=1)\n[161/350] FAIL context_tracking-011 (pred=1, label=2)\n[162/350] FAIL context_tracking-012 (pred=2, label=3)\n[163/350] PASS context_tracking-013 (pred=0, label=0)\n[164/350] PASS context_tracking-014 (pred=1, label=1)\n[165/350] PASS context_tracking-015 (pred=2, label=2)\n[166/350] PASS context_tracking-016 (pred=3, label=3)\n[167/350] FAIL context_tracking-017 (pred=1, label=0)\n[168/350] FAIL context_tracking-018 (pred=0, label=1)\n[169/350] FAIL context_tracking-019 (pred=1, label=2)\n[170/350] PASS context_tracking-020 (pred=3, label=3)\n[171/350] FAIL context_tracking-021 (pred=1, label=0)\n[172/350] FAIL context_tracking-022 (pred=0, label=1)\n[173/350] PASS context_tracking-023 (pred=2, label=2)\n[174/350] FAIL context_tracking-024 (pred=0, label=3)\n[175/350] FAIL context_tracking-025 (pred=1, label=0)\n[176/350] FAIL context_tracking-026 (pred=3, label=1)\n[177/350] FAIL context_tracking-027 (pred=0, label=2)\n[178/350] FAIL context_tracking-028 (pred=0, label=3)\n[179/350] FAIL context_tracking-029 (pred=3, label=0)\n[180/350] PASS context_tracking-030 (pred=1, label=1)\n[181/350] FAIL context_tracking-031 (pred=0, label=2)\n[182/350] FAIL context_tracking-032 (pred=1, label=3)\n[183/350] FAIL context_tracking-033 (pred=1, label=0)\n[184/350] PASS context_tracking-034 (pred=1, label=1)\n[185/350] FAIL context_tracking-035 (pred=0, label=2)\n[186/350] PASS context_tracking-036 (pred=3, label=3)\n[187/350] FAIL context_tracking-037 (pred=2, label=0)\n[188/350] PASS context_tracking-038 (pred=1, label=1)\n[189/350] PASS context_tracking-039 (pred=2, label=2)\n[190/350] PASS context_tracking-040 (pred=3, label=3)\n[191/350] FAIL context_tracking-041 (pred=2, label=0)\n[192/350] FAIL context_tracking-042 (pred=2, label=1)\n[193/350] FAIL context_tracking-043 (pred=1, label=2)\n[194/350] FAIL context_tracking-044 (pred=1, label=3)\n[195/350] FAIL context_tracking-045 (pred=3, label=0)\n[196/350] FAIL context_tracking-046 (pred=3, label=1)\n[197/350] FAIL context_tracking-047 (pred=0, label=2)\n[198/350] PASS context_tracking-048 (pred=3, label=3)\n[199/350] FAIL context_tracking-049 (pred=1, label=0)\n[200/350] FAIL context_tracking-050 (pred=3, label=1)\n[201/350] FAIL quantitative-001 (pred=2, label=0)\n[202/350] FAIL quantitative-002 (pred=2, label=3)\n[203/350] PASS quantitative-003 (pred=2, label=2)\n[204/350] FAIL quantitative-004 (pred=1, label=3)\n[205/350] PASS quantitative-005 (pred=0, label=0)\n[206/350] FAIL quantitative-006 (pred=0, label=1)\n[207/350] FAIL quantitative-007 (pred=3, label=2)\n[208/350] FAIL quantitative-008 (pred=0, label=3)\n[209/350] FAIL quantitative-009 (pred=2, label=0)\n[210/350] PASS quantitative-010 (pred=1, label=1)\n[211/350] PASS quantitative-011 (pred=2, label=2)\n[212/350] PASS quantitative-012 (pred=3, label=3)\n[213/350] FAIL quantitative-013 (pred=3, label=0)\n[214/350] FAIL quantitative-014 (pred=0, label=1)\n[215/350] FAIL quantitative-015 (pred=3, label=2)\n[216/350] FAIL quantitative-016 (pred=2, label=3)\n[217/350] PASS quantitative-017 (pred=0, label=0)\n[218/350] FAIL quantitative-018 (pred=0, label=1)\n[219/350] FAIL quantitative-019 (pred=1, label=2)\n[220/350] PASS quantitative-020 (pred=3, label=3)\n[221/350] FAIL quantitative-021 (pred=2, label=0)\n[222/350] FAIL quantitative-022 (pred=3, label=1)\n[223/350] FAIL quantitative-023 (pred=3, label=2)\n[224/350] FAIL quantitative-024 (pred=0, label=3)\n[225/350] PASS quantitative-025 (pred=0, label=0)\n[226/350] FAIL quantitative-026 (pred=0, label=1)\n[227/350] PASS quantitative-027 (pred=2, label=2)\n[228/350] FAIL quantitative-028 (pred=2, label=3)\n[229/350] FAIL quantitative-029 (pred=3, label=0)\n[230/350] PASS quantitative-030 (pred=1, label=1)\n[231/350] FAIL quantitative-031 (pred=1, label=2)\n[232/350] FAIL quantitative-032 (pred=2, label=3)\n[233/350] FAIL quantitative-033 (pred=3, label=0)\n[234/350] FAIL quantitative-034 (pred=0, label=1)\n[235/350] FAIL quantitative-035 (pred=0, label=2)\n[236/350] FAIL quantitative-036 (pred=1, label=3)\n[237/350] FAIL quantitative-037 (pred=1, label=0)\n[238/350] FAIL quantitative-038 (pred=0, label=1)\n[239/350] FAIL quantitative-039 (pred=3, label=2)\n[240/350] FAIL quantitative-040 (pred=2, label=3)\n[241/350] FAIL quantitative-041 (pred=1, label=0)\n[242/350] FAIL quantitative-042 (pred=0, label=1)\n[243/350] PASS quantitative-043 (pred=2, label=2)\n[244/350] PASS quantitative-044 (pred=3, label=3)\n[245/350] FAIL quantitative-045 (pred=1, label=0)\n[246/350] FAIL quantitative-046 (pred=2, label=1)\n[247/350] FAIL quantitative-047 (pred=3, label=2)\n[248/350] FAIL quantitative-048 (pred=0, label=3)\n[249/350] FAIL quantitative-049 (pred=3, label=0)\n[250/350] PASS quantitative-050 (pred=1, label=1)\n[251/350] PASS logical_reasoning-001 (pred=0, label=0)\n[252/350] FAIL logical_reasoning-002 (pred=0, label=3)\n[253/350] PASS logical_reasoning-003 (pred=2, label=2)\n[254/350] FAIL logical_reasoning-004 (pred=0, label=3)\n[255/350] PASS logical_reasoning-005 (pred=0, label=0)\n[256/350] PASS logical_reasoning-006 (pred=1, label=1)\n[257/350] PASS logical_reasoning-007 (pred=2, label=2)\n[258/350] PASS logical_reasoning-008 (pred=3, label=3)\n[259/350] PASS logical_reasoning-009 (pred=0, label=0)\n[260/350] PASS logical_reasoning-010 (pred=1, label=1)\n[261/350] FAIL logical_reasoning-011 (pred=3, label=2)\n[262/350] FAIL logical_reasoning-012 (pred=2, label=3)\n[263/350] FAIL logical_reasoning-013 (pred=2, label=0)\n[264/350] PASS logical_reasoning-014 (pred=1, label=1)\n[265/350] PASS logical_reasoning-015 (pred=2, label=2)\n[266/350] FAIL logical_reasoning-016 (pred=2, label=3)\n[267/350] FAIL logical_reasoning-017 (pred=1, label=0)\n[268/350] PASS logical_reasoning-018 (pred=1, label=1)\n[269/350] PASS logical_reasoning-019 (pred=2, label=2)\n[270/350] FAIL logical_reasoning-020 (pred=0, label=3)\n[271/350] PASS logical_reasoning-021 (pred=0, label=0)\n[272/350] FAIL logical_reasoning-022 (pred=0, label=1)\n[273/350] FAIL logical_reasoning-023 (pred=0, label=2)\n[274/350] FAIL logical_reasoning-024 (pred=2, label=3)\n[275/350] PASS logical_reasoning-025 (pred=0, label=0)\n[276/350] FAIL logical_reasoning-026 (pred=2, label=1)\n[277/350] FAIL logical_reasoning-027 (pred=0, label=2)\n[278/350] FAIL logical_reasoning-028 (pred=0, label=3)\n[279/350] PASS logical_reasoning-029 (pred=0, label=0)\n[280/350] PASS logical_reasoning-030 (pred=1, label=1)\n[281/350] FAIL logical_reasoning-031 (pred=1, label=2)\n[282/350] FAIL logical_reasoning-032 (pred=2, label=3)\n[283/350] FAIL logical_reasoning-033 (pred=1, label=0)\n[284/350] FAIL logical_reasoning-034 (pred=2, label=1)\n[285/350] FAIL logical_reasoning-035 (pred=0, label=2)\n[286/350] FAIL logical_reasoning-036 (pred=2, label=3)\n[287/350] PASS logical_reasoning-037 (pred=0, label=0)\n[288/350] FAIL logical_reasoning-038 (pred=2, label=1)\n[289/350] FAIL logical_reasoning-039 (pred=0, label=2)\n[290/350] FAIL logical_reasoning-040 (pred=2, label=3)\n[291/350] FAIL logical_reasoning-041 (pred=1, label=0)\n[292/350] PASS logical_reasoning-042 (pred=1, label=1)\n[293/350] PASS logical_reasoning-043 (pred=2, label=2)\n[294/350] FAIL logical_reasoning-044 (pred=0, label=3)\n[295/350] PASS logical_reasoning-045 (pred=0, label=0)\n[296/350] FAIL logical_reasoning-046 (pred=2, label=1)\n[297/350] PASS logical_reasoning-047 (pred=2, label=2)\n[298/350] FAIL logical_reasoning-048 (pred=2, label=3)\n[299/350] FAIL logical_reasoning-049 (pred=2, label=0)\n[300/350] PASS logical_reasoning-050 (pred=1, label=1)\n[301/350] PASS code_completion-001 (pred=0, label=0)\n[302/350] FAIL code_completion-002 (pred=0, label=1)\n[303/350] PASS code_completion-003 (pred=2, label=2)\n[304/350] FAIL code_completion-004 (pred=2, label=3)\n[305/350] PASS code_completion-005 (pred=0, label=0)\n[306/350] FAIL code_completion-006 (pred=0, label=1)\n[307/350] FAIL code_completion-007 (pred=0, label=2)\n[308/350] PASS code_completion-008 (pred=3, label=3)\n[309/350] FAIL code_completion-009 (pred=3, label=0)\n[310/350] FAIL code_completion-010 (pred=2, label=1)\n[311/350] FAIL code_completion-011 (pred=0, label=2)\n[312/350] FAIL code_completion-012 (pred=2, label=3)\n[313/350] PASS code_completion-013 (pred=0, label=0)\n[314/350] PASS code_completion-014 (pred=1, label=1)\n[315/350] FAIL code_completion-015 (pred=3, label=2)\n[316/350] FAIL code_completion-016 (pred=1, label=3)\n[317/350] FAIL code_completion-017 (pred=3, label=0)\n[318/350] FAIL code_completion-018 (pred=0, label=1)\n[319/350] FAIL code_completion-019 (pred=1, label=2)\n[320/350] PASS code_completion-020 (pred=3, label=3)\n[321/350] PASS code_completion-021 (pred=0, label=0)\n[322/350] FAIL code_completion-022 (pred=0, label=1)\n[323/350] PASS code_completion-023 (pred=2, label=2)\n[324/350] FAIL code_completion-024 (pred=1, label=3)\n[325/350] FAIL code_completion-025 (pred=1, label=0)\n[326/350] FAIL code_completion-026 (pred=2, label=1)\n[327/350] FAIL code_completion-027 (pred=0, label=2)\n[328/350] FAIL code_completion-028 (pred=2, label=3)\n[329/350] FAIL code_completion-029 (pred=2, label=0)\n[330/350] FAIL code_completion-030 (pred=3, label=1)\n[331/350] FAIL code_completion-031 (pred=3, label=2)\n[332/350] FAIL code_completion-032 (pred=0, label=3)\n[333/350] FAIL code_completion-033 (pred=1, label=0)\n[334/350] FAIL code_completion-034 (pred=0, label=1)\n[335/350] FAIL code_completion-035 (pred=1, label=2)\n[336/350] FAIL code_completion-036 (pred=0, label=3)\n[337/350] FAIL code_completion-037 (pred=1, label=0)\n[338/350] FAIL code_completion-038 (pred=0, label=1)\n[339/350] FAIL code_completion-039 (pred=1, label=2)\n[340/350] FAIL code_completion-040 (pred=2, label=3)\n[341/350] PASS code_completion-041 (pred=0, label=0)\n[342/350] PASS code_completion-042 (pred=1, label=1)\n[343/350] PASS code_completion-043 (pred=2, label=2)\n[344/350] PASS code_completion-044 (pred=3, label=3)\n[345/350] PASS code_completion-045 (pred=0, label=0)\n[346/350] FAIL code_completion-046 (pred=0, label=1)\n[347/350] FAIL code_completion-047 (pred=0, label=2)\n[348/350] FAIL code_completion-048 (pred=2, label=3)\n[349/350] FAIL code_completion-049 (pred=2, label=0)\n[350/350] FAIL code_completion-050 (pred=3, label=1)\n\nBananaMind Base Bench 1.1\nOverall Elo: 1007\nAccuracy: 187/350 (53.43%)\nWeighted accuracy: 48.92%\nlanguage_completion: Elo 1273 | 45/50 (90.00%) | weighted 90.13%\ncommonsense: Elo 1132 | 38/50 (76.00%) | weighted 73.19%\nworld_knowledge: Elo 1112 | 37/50 (74.00%) | weighted 70.85%\ncontext_tracking: Elo 874 | 18/50 (36.00%) | weighted 33.44%\nquantitative: Elo 840 | 13/50 (26.00%) | weighted 23.66%\nlogical_reasoning: Elo 1031 | 22/50 (44.00%) | weighted 40.75%\ncode_completion: Elo 921 | 14/50 (28.00%) | weighted 27.71%\nReport: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1/report.json\nPredictions: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1/predictions.md\nResult JSON: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1/report.json\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"   elo.scale  elo.prior_rating  elo.prior_weight  summary.accuracy  \\\n0      400.0            1000.0               4.0          0.534286   \n\n   summary.weighted_points  summary.possible_weighted_points  \\\n0                    326.9                          668.1875   \n\n   summary.weighted_accuracy  summary.overall_elo  \\\n0                   0.489234                 1007   \n\n   summary.overall_elo_unrounded  \\\n0                    1007.356924   \n\n   summary.categories.language_completion.accuracy  ...  \\\n0                                              0.9  ...   \n\n   results[340].item_elo  results[341].item_elo  results[342].item_elo  \\\n0                   1200                   1200                   1200   \n\n   results[343].item_elo  results[344].item_elo  results[345].item_elo  \\\n0                   1200                   1200                   1200   \n\n   results[346].item_elo  results[347].item_elo  results[348].item_elo  \\\n0                   1200                   1200                   1200   \n\n   results[349].item_elo  \n0                   1200  \n\n[1 rows x 419 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>elo.scale</th>\n      <th>elo.prior_rating</th>\n      <th>elo.prior_weight</th>\n      <th>summary.accuracy</th>\n      <th>summary.weighted_points</th>\n      <th>summary.possible_weighted_points</th>\n      <th>summary.weighted_accuracy</th>\n      <th>summary.overall_elo</th>\n      <th>summary.overall_elo_unrounded</th>\n      <th>summary.categories.language_completion.accuracy</th>\n      <th>...</th>\n      <th>results[340].item_elo</th>\n      <th>results[341].item_elo</th>\n      <th>results[342].item_elo</th>\n      <th>results[343].item_elo</th>\n      <th>results[344].item_elo</th>\n      <th>results[345].item_elo</th>\n      <th>results[346].item_elo</th>\n      <th>results[347].item_elo</th>\n      <th>results[348].item_elo</th>\n      <th>results[349].item_elo</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>400.0</td>\n      <td>1000.0</td>\n      <td>4.0</td>\n      <td>0.534286</td>\n      <td>326.9</td>\n      <td>668.1875</td>\n      <td>0.489234</td>\n      <td>1007</td>\n      <td>1007.356924</td>\n      <td>0.9</td>\n      <td>...</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n    </tr>\n  </tbody>\n</table>\n<p>1 rows × 419 columns</p>\n</div>"},"metadata":{}}],"execution_count":17},{"id":"fac032c7","cell_type":"markdown","source":"## 9. ArithMark 3.0 — official runner","metadata":{"id":"fac032c7"}},{"id":"e69fb82b","cell_type":"code","source":"def run_arithmark3(batch_size=ARITH3_BATCH):\n    out_dir = ROOT / \"arithmark3\"\n    out_dir.mkdir(parents=True, exist_ok=True)\n\n    script = Path(hf_hub_download(\n        repo_id=ARITH3_REPO,\n        repo_type=\"dataset\",\n        filename=\"bencharithmark-3.py\",\n        revision=ARITH3_REVISION,\n        token=HF_TOKEN,\n    ))\n    data_file = Path(hf_hub_download(\n        repo_id=ARITH3_REPO,\n        repo_type=\"dataset\",\n        filename=\"arithmark-3.jsonl\",\n        revision=ARITH3_REVISION,\n        token=HF_TOKEN,\n    ))\n\n    cmd = [\n        sys.executable,\n        str(script),\n        \"--model\", ACTIVE_MODEL,\n        \"--device\", \"cuda\" if torch.cuda.is_available() else \"cpu\",\n        \"--dtype\", BENCH_DTYPE,\n        \"--batch-size\", str(batch_size),\n        \"--max-context\", str(BENCH_MAX_LENGTH),\n        \"--data-path\", str(data_file),\n        \"--primary-metric\", \"acc_norm\",\n        \"--results-dir\", str(out_dir),\n    ]\n\n    _run(cmd, cwd=out_dir)\n\n    path = newest_json(out_dir)\n    payload = print_json_path(path)\n\n    if payload is not None:\n        flat = flatten_dict(payload)\n        interesting = {\n            k: v for k, v in flat.items()\n            if (\"acc\" in k.lower() or \"correct\" in k.lower() or \"total\" in k.lower())\n            and isinstance(v, (int, float))\n        }\n        if interesting:\n            display(pd.DataFrame([interesting]))\n\n    return {\n        \"raw\": payload,\n        \"path\": str(path) if path else None,\n        \"output_dir\": str(out_dir),\n    }\n\narith_scores = run_arithmark3()\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["6decf21dd8b6432e8e9119776aaecd4f","770bad2022af4be88752ae58f73a688b","b1be3ee818b04e2ca6544069e75dcb61","a8997d8ee6f94441803a3c171a1524cc","22d5d5dfbe824bb491d9c98d234b00a6","f88ee16075ab4d03bc9376ab6d5762d1","2ab67ac614404fbc865e7c396f8ac5d7","c4a9a778f9eb4f5c84747d1f722c8bd0","da0d938e11e8450ab0a1f6efcfbb1c5b","13b1fd360be24a3a93d73e30d6f24033","b1a21592565b4430926b7e9217d583d2","7652e9089b104c9d91988b49e30ab748","5506dc21fc2c4d7896d45c9b75830293","db79212074c4426fa86c1d885be4b341","73bfc6154a5d449cbfd57dae9cab705c","cbfe3fdb0f884880ba1f080ce4232d09","e9a4d884660048c8bd8321563f0ff855","fbdc5adcc5e349df9978d7877956d997","b4a983f8062044328c595abfeef09e6f","56b3d94a40514bf6b7ab6369bae6be71","f538ca606f754965be6359213171709b","5d34eeb0d6b94968be6fa15102cd0e80"]},"id":"e69fb82b","outputId":"69bffb5c-f667-47ab-f4d8-a9f14c1191f2","trusted":true,"jupyter":{"source_hidden":true},"scrolled":true,"execution":{"iopub.status.busy":"2026-10-09T02:14:57.823790Z","iopub.execute_input":"2026-10-09T02:14:57.824018Z","iopub.status.idle":"2026-10-09T02:15:20.344423Z","shell.execute_reply.started":"2026-10-09T02:14:57.823989Z","shell.execute_reply":"2026-10-09T02:15:20.343823Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"bencharithmark-3.py: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"72ab249ff5974b868fec56d1f8aec0b9"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"arithmark-3.jsonl: 0.00B [00:00, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"1cd82048aa0247548e173f7958a96598"}},"metadata":{}},{"name":"stdout","text":"\nCOMMAND: /usr/bin/python3 /root/.cache/huggingface/hub/datasets--AxiomicLabs--Arithmark-3.0/snapshots/6f6e59dd9b7e2c63455f7af7f838f9ecc3d0a746/bencharithmark-3.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --device cuda --dtype float16 --batch-size 32 --max-context 2048 --data-path /root/.cache/huggingface/hub/datasets--AxiomicLabs--Arithmark-3.0/snapshots/6f6e59dd9b7e2c63455f7af7f838f9ecc3d0a746/arithmark-3.jsonl --primary-metric acc_norm --results-dir /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/arithmark3\nLoaded 1000 ArithMark 3.0 examples from /root/.cache/huggingface/hub/datasets--AxiomicLabs--Arithmark-3.0/blobs/237840149650455c4c54ff18f02203e2a76c1d75\nArithMark 3.0 dataset SHA-256: bf8ab1a5193d52cdf0e05ff0b3ca226bdfcf416cb6e75562dcbe72e7e4559435\nUsing device: cuda\n\n====================================================================\n  Loading /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model...\n====================================================================\n  90,148,352 parameters (torch.float16)\n\n  tokenizing:   0%|          | 0/1000 [00:00<?, ?it/s]\n  tokenizing:  14%|█▍        | 144/1000 [00:00<00:00, 1432.81it/s]\n  tokenizing:  30%|██▉       | 295/1000 [00:00<00:00, 1472.79it/s]\n  tokenizing:  45%|████▍     | 449/1000 [00:00<00:00, 1503.32it/s]\n  tokenizing:  60%|██████    | 600/1000 [00:00<00:00, 1497.98it/s]\n  tokenizing:  75%|███████▌  | 752/1000 [00:00<00:00, 1503.05it/s]\n  tokenizing:  90%|█████████ | 905/1000 [00:00<00:00, 1510.25it/s]\n                                                                  \n\n  arithmark-3:   0%|          | 0/32 [00:00<?, ?it/s]\n  arithmark-3:   3%|▎         | 1/32 [00:00<00:21,  1.47it/s]\n  arithmark-3:   6%|▋         | 2/32 [00:00<00:11,  2.53it/s]\n  arithmark-3:   9%|▉         | 3/32 [00:01<00:08,  3.23it/s]\n  arithmark-3:  12%|█▎        | 4/32 [00:01<00:07,  3.71it/s]\n  arithmark-3:  16%|█▌        | 5/32 [00:01<00:06,  4.02it/s]\n  arithmark-3:  19%|█▉        | 6/32 [00:01<00:06,  4.20it/s]\n  arithmark-3:  22%|██▏       | 7/32 [00:01<00:05,  4.27it/s]\n  arithmark-3:  25%|██▌       | 8/32 [00:02<00:05,  4.29it/s]\n  arithmark-3:  28%|██▊       | 9/32 [00:02<00:05,  4.29it/s]\n  arithmark-3:  31%|███▏      | 10/32 [00:02<00:05,  4.28it/s]\n  arithmark-3:  34%|███▍      | 11/32 [00:02<00:04,  4.23it/s]\n  arithmark-3:  38%|███▊      | 12/32 [00:03<00:04,  4.18it/s]\n  arithmark-3:  41%|████      | 13/32 [00:03<00:04,  4.12it/s]\n  arithmark-3:  44%|████▍     | 14/32 [00:03<00:04,  4.09it/s]\n  arithmark-3:  47%|████▋     | 15/32 [00:03<00:04,  4.05it/s]\n  arithmark-3:  50%|█████     | 16/32 [00:04<00:04,  4.00it/s]\n  arithmark-3:  53%|█████▎    | 17/32 [00:04<00:03,  3.94it/s]\n  arithmark-3:  56%|█████▋    | 18/32 [00:04<00:03,  3.87it/s]\n  arithmark-3:  59%|█████▉    | 19/32 [00:04<00:03,  3.79it/s]\n  arithmark-3:  62%|██████▎   | 20/32 [00:05<00:03,  3.72it/s]\n  arithmark-3:  66%|██████▌   | 21/32 [00:05<00:03,  3.64it/s]\n  arithmark-3:  69%|██████▉   | 22/32 [00:05<00:02,  3.59it/s]\n  arithmark-3:  72%|███████▏  | 23/32 [00:06<00:02,  3.53it/s]\n  arithmark-3:  75%|███████▌  | 24/32 [00:06<00:02,  3.45it/s]\n  arithmark-3:  78%|███████▊  | 25/32 [00:06<00:02,  3.39it/s]\n  arithmark-3:  81%|████████▏ | 26/32 [00:07<00:01,  3.31it/s]\n  arithmark-3:  84%|████████▍ | 27/32 [00:07<00:01,  3.24it/s]\n  arithmark-3:  88%|████████▊ | 28/32 [00:07<00:01,  3.15it/s]\n  arithmark-3:  91%|█████████ | 29/32 [00:08<00:00,  3.01it/s]\n  arithmark-3:  94%|█████████▍| 30/32 [00:08<00:00,  2.82it/s]\n  arithmark-3:  97%|█████████▋| 31/32 [00:08<00:00,  2.64it/s]\n  arithmark-3: 100%|██████████| 32/32 [00:09<00:00,  3.31it/s]\n  arithmark-3: 100%|██████████| 32/32 [00:09<00:00,  3.55it/s]\n  arithmark-3: raw 44.90% (449/1000)  normalized 44.80% (448/1000)\n  speed: 110.8 examples/s  (tokenize 0.67s, evaluate 9.03s)\n\n  Category                                                             N       Raw   Normalized\n  ----------------------------------------------------------------------------------------------\n  elementary_school_math_continuation::addition::grades_1_2::easy    128    33.59%       33.59%\n  elementary_school_math_continuation::comparison::grades_2_3::medium    44    38.64%       38.64%\n  elementary_school_math_continuation::comparison_difference::grades_2_3::medium    48    39.58%       39.58%\n  elementary_school_math_continuation::data::grades_2_3::easy         43    34.88%       34.88%\n  elementary_school_math_continuation::division::grades_3_4::medium    54    48.15%       48.15%\n  elementary_school_math_continuation::fractions_counting::grades_3_4::medium    50    20.00%       18.00%\n  elementary_school_math_continuation::geometry_area::grades_4_5::medium    52    84.62%       86.54%\n  elementary_school_math_continuation::geometry_perimeter::grades_4_5::medium    45    66.67%       66.67%\n  elementary_school_math_continuation::measurement::grades_2_3::easy    76    36.84%       36.84%\n  elementary_school_math_continuation::money::grades_3_4::medium      64    35.94%       34.38%\n  elementary_school_math_continuation::multiplication::grades_3_4::medium    74    77.03%       75.68%\n  elementary_school_math_continuation::patterns::grades_3_4::medium    53    37.74%       37.74%\n  elementary_school_math_continuation::subtraction::grades_1_2::easy   117    30.77%       30.77%\n  elementary_school_math_continuation::time::grades_2_3::easy         55   100.00%      100.00%\n  elementary_school_math_continuation::two_step_add_subtract::grades_2_3::medium    46    26.09%       26.09%\n  elementary_school_math_continuation::two_step_addition::grades_2_3::medium    19    21.05%       21.05%\n  elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium    32    31.25%       34.38%\n\n====================================================================\n  /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model (90,148,352 params) RESULTS\n====================================================================\n  Raw continuation accuracy        44.90%\n  Length-normalized accuracy       44.80%\n  Primary (acc_norm)           44.80%\n====================================================================\nResults saved to /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/arithmark3/_kaggle_working_void_eval_appvoid--void-byte_92c601560cc2179e3bfc33e067fdd3bf9254e163__void_model_arithmark-3_results.json\nResult JSON: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/arithmark3/_kaggle_working_void_eval_appvoid--void-byte_92c601560cc2179e3bfc33e067fdd3bf9254e163__void_model_arithmark-3_results.json\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"   results.arithmark-3.acc  results.arithmark-3.acc_norm  \\\n0                     44.9                          44.8   \n\n   results.arithmark-3.raw_correct  results.arithmark-3.norm_correct  \\\n0                              449                               448   \n\n   results.arithmark-3.total  \\\n0                       1000   \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc  \\\n0                                           33.59375                                                    \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc_norm  \\\n0                                           33.59375                                                         \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.raw_correct  \\\n0                                                 43                                                            \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.norm_correct  \\\n0                                                 43                                                             \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.total  \\\n0                                                128                                                      \n\n   ...  \\\n0  ...   \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.acc_norm  \\\n0                                          21.052632                                                                    \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.raw_correct  \\\n0                                                  4                                                                       \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.norm_correct  \\\n0                                                  4                                                                        \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.total  \\\n0                                                 19                                                                 \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc  \\\n0                                              31.25                                                                  \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc_norm  \\\n0                                             34.375                                                                       \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.raw_correct  \\\n0                                                 10                                                                          \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.norm_correct  \\\n0                                                 11                                                                           \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.total  \\\n0                                                 32                                                                    \n\n   results.arithmark-3.primary_acc  \n0                             44.8  \n\n[1 rows x 91 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>results.arithmark-3.acc</th>\n      <th>results.arithmark-3.acc_norm</th>\n      <th>results.arithmark-3.raw_correct</th>\n      <th>results.arithmark-3.norm_correct</th>\n      <th>results.arithmark-3.total</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc_norm</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.raw_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.norm_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.total</th>\n      <th>...</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.acc_norm</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.raw_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.norm_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.total</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc_norm</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.raw_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.norm_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.total</th>\n      <th>results.arithmark-3.primary_acc</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>44.9</td>\n      <td>44.8</td>\n      <td>449</td>\n      <td>448</td>\n      <td>1000</td>\n      <td>33.59375</td>\n      <td>33.59375</td>\n      <td>43</td>\n      <td>43</td>\n      <td>128</td>\n      <td>...</td>\n      <td>21.052632</td>\n      <td>4</td>\n      <td>4</td>\n      <td>19</td>\n      <td>31.25</td>\n      <td>34.375</td>\n      <td>10</td>\n      <td>11</td>\n      <td>32</td>\n      <td>44.8</td>\n    </tr>\n  </tbody>\n</table>\n<p>1 rows × 91 columns</p>\n</div>"},"metadata":{}}],"execution_count":18},{"id":"58e727e2","cell_type":"markdown","source":"## 10. Unified suite and Intelligence Index","metadata":{"id":"58e727e2"}},{"id":"0e3bfb29","cell_type":"code","source":"def _deep_numeric(payload, names):\n    if payload is None:\n        return None\n\n    flat = flatten_dict(payload)\n    lowered_names = [x.lower() for x in names]\n\n    for k, v in flat.items():\n        if not isinstance(v, (int, float)):\n            continue\n        kl = k.lower()\n        if any(kl.endswith(name) for name in lowered_names):\n            return float(v)\n\n    for k, v in flat.items():\n        if not isinstance(v, (int, float)):\n            continue\n        kl = k.lower()\n        if any(name in kl for name in lowered_names):\n            return float(v)\n\n    return None\n\ndef extract_arithmark3_acc_norm(payload):\n    value = _deep_numeric(payload, [\"acc_norm\", \"normalized_accuracy\"])\n    if value is None:\n        return None\n    return value * 100 if abs(value) <= 1 else value\n\ndef extract_bananamind(payload):\n    return {\n        \"bananamind_elo\": _deep_numeric(\n            payload, [\"overall_elo\", \"overall.elo\", \"elo\"]\n        ),\n        \"bananamind_accuracy\": _deep_numeric(\n            payload, [\"overall_accuracy\", \"raw_accuracy\", \"accuracy\"]\n        ),\n        \"bananamind_weighted_accuracy\": _deep_numeric(\n            payload, [\"weighted_accuracy\", \"weighted_acc\"]\n        ),\n    }\n\ndef chance_normalize(score, chance):\n    return 100.0 * (score - chance) / (100.0 - chance)\n\ndef intelligence_index(\n    hellaswag,\n    arc_easy,\n    arc_challenge,\n    piqa,\n    arithmark3,\n):\n    arc_mean = (arc_easy + arc_challenge) / 2.0\n    return (\n        chance_normalize(hellaswag, 25.0)\n        + chance_normalize(arc_mean, 25.0)\n        + chance_normalize(piqa, 50.0)\n        + 0.65 * chance_normalize(arithmark3, 25.0)\n    ) / 3.65\n\ndef run_full_suite(run_banana=True, run_arith=True):\n    print(\"=\" * 80)\n    print(\"VOID FULL BENCHMARK SUITE\")\n    print(\"=\" * 80)\n    print(\"Pinned revision:\", RESOLVED_MODEL_REVISION or MODEL_REVISION)\n    print(\"Benchmark max length:\", BENCH_MAX_LENGTH)\n\n    std = run_open_slm()\n    banana = run_bananamind() if run_banana else None\n    arith = run_arithmark3() if run_arith else None\n\n    row = dict(std[\"summary\"])\n\n    if arith is not None:\n        row[\"arithmark3\"] = extract_arithmark3_acc_norm(arith[\"raw\"])\n\n    if banana is not None:\n        row.update(extract_bananamind(banana[\"raw\"]))\n\n    required = [\n        \"hellaswag\",\n        \"arc_easy\",\n        \"arc_challenge\",\n        \"piqa\",\n        \"arithmark3\",\n    ]\n    if all(row.get(k) is not None for k in required):\n        row[\"intelligence_index\"] = intelligence_index(\n            row[\"hellaswag\"],\n            row[\"arc_easy\"],\n            row[\"arc_challenge\"],\n            row[\"piqa\"],\n            row[\"arithmark3\"],\n        )\n\n    row[\"benchmark_max_length\"] = BENCH_MAX_LENGTH\n    unified_path = ROOT / \"void_unified_summary.json\"\n    unified_path.write_text(json.dumps(row, indent=2), encoding=\"utf-8\")\n\n    print(\"\\nUnified summary\")\n    display(pd.DataFrame([row]))\n    print(\"Saved:\", unified_path)\n    return row\n\n# Full evaluation:\nvoid_scores = run_full_suite()\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"0e3bfb29","outputId":"fcfc4f75-91d7-4b80-82c1-013394bd7d1f","trusted":true,"jupyter":{"source_hidden":true},"scrolled":true,"execution":{"iopub.status.busy":"2026-10-09T02:15:20.345552Z","iopub.execute_input":"2026-10-09T02:15:20.345948Z","iopub.status.idle":"2026-10-09T02:27:54.520153Z","shell.execute_reply.started":"2026-10-09T02:15:20.345915Z","shell.execute_reply":"2026-10-09T02:27:54.519546Z"}},"outputs":[{"name":"stdout","text":"================================================================================\nVOID FULL BENCHMARK SUITE\n================================================================================\nPinned revision: 92c601560cc2179e3bfc33e067fdd3bf9254e163\nBenchmark max length: 2048\nFreeing interactive Void model before benchmark subprocess...\n\n================================================================================\nVOID — hellaswag\n================================================================================\n\n[hellaswag] trying fixed batch=512\n\nCOMMAND: /usr/bin/python3 /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_run_void_lm_eval.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --revision 92c601560cc2179e3bfc33e067fdd3bf9254e163 --task hellaswag --device cuda:0 --dtype float16 --batch-size 512 --max-length 2048 --long-cont-stride 256 --out /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/open_slm/hellaswag/results.json\ntask = hellaswag batch = 512 max_length = 2048 long_cont_stride = 256\n\nGenerating train split:   0%|          | 0/39905 [00:00<?, ? examples/s]\nGenerating train split:   3%|▎         | 1000/39905 [00:00<00:03, 9824.01 examples/s]\nGenerating train split:  88%|████████▊ | 35000/39905 [00:00<00:00, 200914.67 examples/s]\nGenerating train split: 100%|██████████| 39905/39905 [00:00<00:00, 177156.46 examples/s]\n\nGenerating test split:   0%|          | 0/10003 [00:00<?, ? examples/s]\nGenerating test split: 100%|██████████| 10003/10003 [00:00<00:00, 249016.96 examples/s]\n\nGenerating validation split:   0%|          | 0/10042 [00:00<?, ? examples/s]\nGenerating validation split: 100%|██████████| 10042/10042 [00:00<00:00, 280075.81 examples/s]\n\nMap:   0%|          | 0/39905 [00:00<?, ? examples/s]\nMap:   3%|▎         | 1000/39905 [00:00<00:06, 5832.45 examples/s]\nMap:   5%|▌         | 2000/39905 [00:00<00:05, 7523.19 examples/s]\nMap:   8%|▊         | 3000/39905 [00:00<00:04, 8313.25 examples/s]\nMap:  10%|█         | 4000/39905 [00:00<00:04, 8788.65 examples/s]\nMap:  13%|█▎        | 5000/39905 [00:00<00:03, 8970.35 examples/s]\nMap:  15%|█▌        | 6000/39905 [00:00<00:03, 9136.27 examples/s]\nMap:  18%|█▊        | 7000/39905 [00:00<00:03, 9273.80 examples/s]\nMap:  20%|██        | 8000/39905 [00:00<00:03, 9411.48 examples/s]\nMap:  23%|██▎       | 9000/39905 [00:01<00:03, 9417.07 examples/s]\nMap:  25%|██▌       | 10000/39905 [00:01<00:03, 9470.74 examples/s]\nMap:  28%|██▊       | 11000/39905 [00:01<00:03, 9502.02 examples/s]\nMap:  30%|███       | 12000/39905 [00:01<00:03, 9011.83 examples/s]\nMap:  33%|███▎      | 13000/39905 [00:01<00:02, 9159.53 examples/s]\nMap:  35%|███▌      | 14000/39905 [00:01<00:02, 9059.46 examples/s]\nMap:  38%|███▊      | 15001/39905 [00:01<00:03, 8162.43 examples/s]\nMap:  40%|████      | 15992/39905 [00:01<00:02, 8615.03 examples/s]\nMap:  43%|████▎     | 17115/39905 [00:01<00:02, 8209.11 examples/s]\nMap:  45%|████▌     | 18000/39905 [00:02<00:02, 8277.31 examples/s]\nMap:  48%|████▊     | 19000/39905 [00:02<00:05, 4139.08 examples/s]\nMap:  50%|████▉     | 19952/39905 [00:02<00:04, 4954.68 examples/s]\nMap:  52%|█████▏    | 20771/39905 [00:02<00:03, 5529.20 examples/s]\nMap:  54%|█████▍    | 21582/39905 [00:02<00:03, 6050.94 examples/s]\nMap:  56%|█████▋    | 22493/39905 [00:03<00:02, 6587.77 examples/s]\nMap:  58%|█████▊    | 23302/39905 [00:03<00:02, 6945.65 examples/s]\nMap:  62%|██████▏   | 24566/39905 [00:03<00:02, 7444.55 examples/s]\nMap:  64%|██████▍   | 25487/39905 [00:03<00:01, 7705.98 examples/s]\nMap:  67%|██████▋   | 26794/39905 [00:03<00:01, 8036.72 examples/s]\nMap:  70%|███████   | 28000/39905 [00:03<00:01, 7898.99 examples/s]\nMap:  73%|███████▎  | 28950/39905 [00:03<00:01, 8271.03 examples/s]\nMap:  75%|███████▌  | 30065/39905 [00:03<00:01, 7988.83 examples/s]\nMap:  78%|███████▊  | 31000/39905 [00:04<00:01, 8115.60 examples/s]\nMap:  80%|████████  | 31947/39905 [00:04<00:00, 8454.37 examples/s]\nMap:  83%|████████▎ | 33061/39905 [00:04<00:00, 8092.62 examples/s]\nMap:  85%|████████▌ | 34000/39905 [00:04<00:00, 8174.23 examples/s]\nMap:  88%|████████▊ | 34993/39905 [00:04<00:00, 8624.30 examples/s]\nMap:  91%|█████████ | 36128/39905 [00:04<00:00, 8236.83 examples/s]\nMap:  93%|█████████▎| 37000/39905 [00:04<00:00, 8273.35 examples/s]\nMap:  95%|█████████▌| 37979/39905 [00:04<00:00, 8669.59 examples/s]\nMap:  98%|█████████▊| 39136/39905 [00:04<00:00, 8314.87 examples/s]\nMap: 100%|██████████| 39905/39905 [00:05<00:00, 7782.87 examples/s]\n\nMap:   0%|          | 0/10042 [00:00<?, ? examples/s]\nMap:  10%|▉         | 1000/10042 [00:00<00:01, 8320.22 examples/s]\nMap:  20%|█▉        | 2000/10042 [00:00<00:00, 8814.46 examples/s]\nMap:  30%|██▉       | 2980/10042 [00:00<00:00, 9228.45 examples/s]\nMap:  41%|████      | 4091/10042 [00:00<00:00, 8356.37 examples/s]\nMap:  50%|████▉     | 5000/10042 [00:00<00:00, 8352.11 examples/s]\nMap:  59%|█████▉    | 5949/10042 [00:00<00:00, 8694.91 examples/s]\nMap:  71%|███████   | 7102/10042 [00:00<00:00, 8295.89 examples/s]\nMap:  80%|███████▉  | 8000/10042 [00:00<00:00, 8231.02 examples/s]\nMap:  89%|████████▉ | 8942/10042 [00:01<00:00, 8555.55 examples/s]\nMap: 100%|██████████| 10042/10042 [00:01<00:00, 8173.47 examples/s]\nMap: 100%|██████████| 10042/10042 [00:01<00:00, 8361.75 examples/s]\nWARNING:lm_eval.evaluator:Overwriting default num_fewshot of hellaswag from None to 0\n\n  0%|          | 0/10042 [00:00<?, ?it/s]\n  1%|          | 121/10042 [00:00<00:08, 1204.03it/s]\n  2%|▏         | 242/10042 [00:00<00:08, 1098.46it/s]\n  4%|▎         | 353/10042 [00:00<00:08, 1098.58it/s]\n  5%|▍         | 476/10042 [00:00<00:08, 1147.77it/s]\n  6%|▌         | 599/10042 [00:00<00:08, 1176.04it/s]\n  7%|▋         | 725/10042 [00:00<00:07, 1201.11it/s]\n  8%|▊         | 849/10042 [00:00<00:07, 1213.31it/s]\n 10%|▉         | 972/10042 [00:00<00:07, 1217.69it/s]\n 11%|█         | 1098/10042 [00:00<00:07, 1228.26it/s]\n 12%|█▏        | 1224/10042 [00:01<00:07, 1236.26it/s]\n 13%|█▎        | 1348/10042 [00:01<00:07, 1234.31it/s]\n 15%|█▍        | 1473/10042 [00:01<00:06, 1238.02it/s]\n 16%|█▌        | 1597/10042 [00:01<00:06, 1232.27it/s]\n 17%|█▋        | 1721/10042 [00:01<00:06, 1233.44it/s]\n 18%|█▊        | 1845/10042 [00:01<00:06, 1233.49it/s]\n 20%|█▉        | 1971/10042 [00:01<00:06, 1240.95it/s]\n 21%|██        | 2096/10042 [00:01<00:06, 1233.79it/s]\n 22%|██▏       | 2220/10042 [00:01<00:06, 1226.22it/s]\n 23%|██▎       | 2346/10042 [00:01<00:06, 1235.54it/s]\n 25%|██▍       | 2472/10042 [00:02<00:06, 1240.86it/s]\n 26%|██▌       | 2597/10042 [00:02<00:06, 1235.31it/s]\n 27%|██▋       | 2721/10042 [00:02<00:05, 1234.28it/s]\n 28%|██▊       | 2845/10042 [00:02<00:05, 1229.89it/s]\n 30%|██▉       | 2968/10042 [00:02<00:05, 1227.37it/s]\n 31%|███       | 3093/10042 [00:02<00:05, 1231.85it/s]\n 32%|███▏      | 3219/10042 [00:02<00:05, 1238.39it/s]\n 33%|███▎      | 3343/10042 [00:02<00:05, 1227.18it/s]\n 35%|███▍      | 3468/10042 [00:02<00:05, 1233.43it/s]\n 36%|███▌      | 3592/10042 [00:02<00:05, 1226.13it/s]\n 37%|███▋      | 3717/10042 [00:03<00:05, 1232.21it/s]\n 38%|███▊      | 3841/10042 [00:03<00:05, 1232.60it/s]\n 39%|███▉      | 3966/10042 [00:03<00:04, 1237.20it/s]\n 41%|████      | 4091/10042 [00:03<00:04, 1238.82it/s]\n 42%|████▏     | 4217/10042 [00:03<00:04, 1244.16it/s]\n 43%|████▎     | 4342/10042 [00:03<00:04, 1240.21it/s]\n 44%|████▍     | 4468/10042 [00:03<00:04, 1245.08it/s]\n 46%|████▌     | 4593/10042 [00:03<00:04, 1239.84it/s]\n 47%|████▋     | 4718/10042 [00:03<00:04, 1241.83it/s]\n 48%|████▊     | 4845/10042 [00:03<00:04, 1248.76it/s]\n 49%|████▉     | 4970/10042 [00:04<00:04, 1237.28it/s]\n 51%|█████     | 5094/10042 [00:04<00:04, 1230.73it/s]\n 52%|█████▏    | 5218/10042 [00:04<00:03, 1231.85it/s]\n 53%|█████▎    | 5342/10042 [00:04<00:03, 1231.73it/s]\n 54%|█████▍    | 5468/10042 [00:04<00:03, 1238.89it/s]\n 56%|█████▌    | 5592/10042 [00:04<00:03, 1233.13it/s]\n 57%|█████▋    | 5716/10042 [00:04<00:03, 1228.44it/s]\n 58%|█████▊    | 5839/10042 [00:04<00:03, 1225.31it/s]\n 59%|█████▉    | 5963/10042 [00:04<00:03, 1226.35it/s]\n 61%|██████    | 6086/10042 [00:04<00:03, 1224.48it/s]\n 62%|██████▏   | 6210/10042 [00:05<00:03, 1227.56it/s]\n 63%|██████▎   | 6333/10042 [00:05<00:03, 1221.16it/s]\n 64%|██████▍   | 6456/10042 [00:05<00:02, 1221.42it/s]\n 66%|██████▌   | 6581/10042 [00:05<00:02, 1227.43it/s]\n 67%|██████▋   | 6704/10042 [00:05<00:02, 1226.34it/s]\n 68%|██████▊   | 6827/10042 [00:05<00:02, 1226.10it/s]\n 69%|██████▉   | 6952/10042 [00:05<00:02, 1230.24it/s]\n 70%|███████   | 7076/10042 [00:05<00:02, 1222.58it/s]\n 72%|███████▏  | 7199/10042 [00:05<00:02, 1224.56it/s]\n 73%|███████▎  | 7324/10042 [00:05<00:02, 1230.10it/s]\n 74%|███████▍  | 7448/10042 [00:06<00:02, 1231.00it/s]\n 75%|███████▌  | 7572/10042 [00:06<00:02, 1228.34it/s]\n 77%|███████▋  | 7696/10042 [00:06<00:01, 1230.96it/s]\n 78%|███████▊  | 7822/10042 [00:06<00:01, 1239.31it/s]\n 79%|███████▉  | 7947/10042 [00:06<00:01, 1241.95it/s]\n 80%|████████  | 8072/10042 [00:06<00:01, 1234.20it/s]\n 82%|████████▏ | 8196/10042 [00:06<00:01, 1235.85it/s]\n 83%|████████▎ | 8320/10042 [00:06<00:01, 1203.00it/s]\n 84%|████████▍ | 8446/10042 [00:06<00:01, 1218.76it/s]\n 85%|████████▌ | 8569/10042 [00:06<00:01, 1213.66it/s]\n 87%|████████▋ | 8694/10042 [00:07<00:01, 1222.77it/s]\n 88%|████████▊ | 8817/10042 [00:07<00:01, 1214.32it/s]\n 89%|████████▉ | 8939/10042 [00:07<00:00, 1210.39it/s]\n 90%|█████████ | 9062/10042 [00:07<00:00, 1213.76it/s]\n 91%|█████████▏| 9185/10042 [00:07<00:00, 1217.27it/s]\n 93%|█████████▎| 9310/10042 [00:07<00:00, 1224.01it/s]\n 94%|█████████▍| 9433/10042 [00:07<00:00, 1224.68it/s]\n 95%|█████████▌| 9556/10042 [00:07<00:00, 1218.01it/s]\n 96%|█████████▋| 9679/10042 [00:07<00:00, 1221.21it/s]\n 98%|█████████▊| 9806/10042 [00:07<00:00, 1235.68it/s]\n 99%|█████████▉| 9937/10042 [00:08<00:00, 1256.73it/s]\n100%|██████████| 10042/10042 [00:08<00:00, 1227.08it/s]\n\nTokenizing inputs:   0%|          | 0/40168 [00:00<?, ?it/s]\nTokenizing inputs:   1%|          | 211/40168 [00:00<00:18, 2109.05it/s]\nTokenizing inputs:   1%|          | 422/40168 [00:00<00:19, 2008.56it/s]\nTokenizing inputs:   2%|▏         | 624/40168 [00:00<00:19, 2003.66it/s]\nTokenizing inputs:   2%|▏         | 825/40168 [00:00<00:19, 1987.22it/s]\nTokenizing inputs:   3%|▎         | 1024/40168 [00:00<00:20, 1939.11it/s]\nTokenizing inputs:   3%|▎         | 1233/40168 [00:00<00:19, 1986.53it/s]\nTokenizing inputs:   4%|▎         | 1442/40168 [00:00<00:19, 2019.31it/s]\nTokenizing inputs:   4%|▍         | 1646/40168 [00:00<00:19, 2024.66it/s]\nTokenizing inputs:   5%|▍         | 1876/40168 [00:00<00:18, 2108.89it/s]\nTokenizing inputs:   5%|▌         | 2096/40168 [00:01<00:17, 2136.52it/s]\nTokenizing inputs:   6%|▌         | 2314/40168 [00:01<00:17, 2148.44it/s]\nTokenizing inputs:   6%|▋         | 2529/40168 [00:01<00:17, 2097.29it/s]\nTokenizing inputs:   7%|▋         | 2740/40168 [00:01<00:18, 2079.25it/s]\nTokenizing inputs:   7%|▋         | 2949/40168 [00:01<00:18, 2067.71it/s]\nTokenizing inputs:   8%|▊         | 3156/40168 [00:01<00:18, 2053.11it/s]\nTokenizing inputs:   8%|▊         | 3367/40168 [00:01<00:17, 2068.63it/s]\nTokenizing inputs:   9%|▉         | 3574/40168 [00:01<00:18, 2019.55it/s]\nTokenizing inputs:   9%|▉         | 3777/40168 [00:01<00:18, 2019.92it/s]\nTokenizing inputs:  10%|▉         | 3981/40168 [00:01<00:17, 2023.12it/s]\nTokenizing inputs:  10%|█         | 4184/40168 [00:02<00:18, 1903.87it/s]\nTokenizing inputs:  11%|█         | 4415/40168 [00:02<00:17, 2019.29it/s]\nTokenizing inputs:  12%|█▏        | 4625/40168 [00:02<00:17, 2041.44it/s]\nTokenizing inputs:  12%|█▏        | 4839/40168 [00:02<00:17, 2070.12it/s]\nTokenizing inputs:  13%|█▎        | 5058/40168 [00:02<00:16, 2103.20it/s]\nTokenizing inputs:  13%|█▎        | 5270/40168 [00:02<00:16, 2075.37it/s]\nTokenizing inputs:  14%|█▎        | 5479/40168 [00:02<00:16, 2078.78it/s]\nTokenizing inputs:  14%|█▍        | 5697/40168 [00:02<00:16, 2107.08it/s]\nTokenizing inputs:  15%|█▍        | 5911/40168 [00:02<00:16, 2116.62it/s]\nTokenizing inputs:  15%|█▌        | 6136/40168 [00:02<00:15, 2154.57it/s]\nTokenizing inputs:  16%|█▌        | 6352/40168 [00:03<00:15, 2114.07it/s]\nTokenizing inputs:  16%|█▋        | 6564/40168 [00:03<00:15, 2107.92it/s]\nTokenizing inputs:  17%|█▋        | 6782/40168 [00:03<00:15, 2127.61it/s]\nTokenizing inputs:  17%|█▋        | 7001/40168 [00:03<00:15, 2145.09it/s]\nTokenizing inputs:  18%|█▊        | 7216/40168 [00:03<00:15, 2135.92it/s]\nTokenizing inputs:  19%|█▊        | 7438/40168 [00:03<00:15, 2159.39it/s]\nTokenizing inputs:  19%|█▉        | 7657/40168 [00:03<00:15, 2166.81it/s]\nTokenizing inputs:  20%|█▉        | 7874/40168 [00:03<00:15, 2148.34it/s]\nTokenizing inputs:  20%|██        | 8090/40168 [00:03<00:14, 2148.26it/s]\nTokenizing inputs:  21%|██        | 8305/40168 [00:03<00:15, 2088.99it/s]\nTokenizing inputs:  21%|██        | 8515/40168 [00:04<00:15, 2080.89it/s]\nTokenizing inputs:  22%|██▏       | 8724/40168 [00:04<00:15, 2060.23it/s]\nTokenizing inputs:  22%|██▏       | 8940/40168 [00:04<00:14, 2087.64it/s]\nTokenizing inputs:  23%|██▎       | 9151/40168 [00:04<00:14, 2092.71it/s]\nTokenizing inputs:  23%|██▎       | 9373/40168 [00:04<00:14, 2129.85it/s]\nTokenizing inputs:  24%|██▍       | 9601/40168 [00:04<00:14, 2173.51it/s]\nTokenizing inputs:  24%|██▍       | 9832/40168 [00:04<00:13, 2213.31it/s]\nTokenizing inputs:  25%|██▌       | 10054/40168 [00:04<00:13, 2203.46it/s]\nTokenizing inputs:  26%|██▌       | 10275/40168 [00:04<00:13, 2203.90it/s]\nTokenizing inputs:  26%|██▌       | 10496/40168 [00:05<00:13, 2154.89it/s]\nTokenizing inputs:  27%|██▋       | 10712/40168 [00:05<00:14, 2080.83it/s]\nTokenizing inputs:  27%|██▋       | 10921/40168 [00:05<00:14, 2036.17it/s]\nTokenizing inputs:  28%|██▊       | 11136/40168 [00:05<00:14, 2067.44it/s]\nTokenizing inputs:  28%|██▊       | 11356/40168 [00:05<00:13, 2104.20it/s]\nTokenizing inputs:  29%|██▉       | 11567/40168 [00:05<00:13, 2092.58it/s]\nTokenizing inputs:  29%|██▉       | 11790/40168 [00:05<00:13, 2131.25it/s]\nTokenizing inputs:  30%|██▉       | 12004/40168 [00:05<00:13, 2121.99it/s]\nTokenizing inputs:  30%|███       | 12238/40168 [00:05<00:12, 2185.89it/s]\nTokenizing inputs:  31%|███       | 12462/40168 [00:05<00:12, 2200.86it/s]\nTokenizing inputs:  32%|███▏      | 12683/40168 [00:06<00:12, 2138.87it/s]\nTokenizing inputs:  32%|███▏      | 12898/40168 [00:06<00:12, 2111.75it/s]\nTokenizing inputs:  33%|███▎      | 13110/40168 [00:06<00:14, 1840.24it/s]\nTokenizing inputs:  33%|███▎      | 13301/40168 [00:06<00:16, 1584.44it/s]\nTokenizing inputs:  34%|███▎      | 13469/40168 [00:06<00:18, 1426.79it/s]\nTokenizing inputs:  34%|███▍      | 13620/40168 [00:06<00:19, 1348.17it/s]\nTokenizing inputs:  34%|███▍      | 13761/40168 [00:06<00:20, 1292.08it/s]\nTokenizing inputs:  35%|███▍      | 13894/40168 [00:06<00:21, 1249.93it/s]\nTokenizing inputs:  35%|███▍      | 14021/40168 [00:07<00:21, 1200.97it/s]\nTokenizing inputs:  35%|███▌      | 14143/40168 [00:07<00:22, 1176.18it/s]\nTokenizing inputs:  36%|███▌      | 14262/40168 [00:07<00:22, 1151.92it/s]\nTokenizing inputs:  36%|███▌      | 14378/40168 [00:07<00:22, 1150.38it/s]\nTokenizing inputs:  36%|███▌      | 14494/40168 [00:07<00:22, 1126.16it/s]\nTokenizing inputs:  36%|███▋      | 14607/40168 [00:07<00:22, 1124.55it/s]\nTokenizing inputs:  37%|███▋      | 14720/40168 [00:07<00:22, 1114.50it/s]\nTokenizing inputs:  37%|███▋      | 14832/40168 [00:07<00:23, 1093.76it/s]\nTokenizing inputs:  37%|███▋      | 14945/40168 [00:07<00:22, 1102.54it/s]\nTokenizing inputs:  37%|███▋      | 15056/40168 [00:08<00:58, 430.55it/s] \nTokenizing inputs:  38%|███▊      | 15172/40168 [00:08<00:47, 531.76it/s]\nTokenizing inputs:  38%|███▊      | 15283/40168 [00:08<00:39, 626.55it/s]\nTokenizing inputs:  38%|███▊      | 15392/40168 [00:08<00:34, 714.25it/s]\nTokenizing inputs:  39%|███▊      | 15502/40168 [00:08<00:31, 795.67it/s]\nTokenizing inputs:  39%|███▉      | 15608/40168 [00:09<00:28, 856.27it/s]\nTokenizing inputs:  39%|███▉      | 15717/40168 [00:09<00:26, 913.55it/s]\nTokenizing inputs:  39%|███▉      | 15828/40168 [00:09<00:25, 963.44it/s]\nTokenizing inputs:  40%|███▉      | 15939/40168 [00:09<00:24, 1002.92it/s]\nTokenizing inputs:  40%|███▉      | 16049/40168 [00:09<00:23, 1028.19it/s]\nTokenizing inputs:  40%|████      | 16158/40168 [00:09<00:23, 1039.41it/s]\nTokenizing inputs:  40%|████      | 16268/40168 [00:09<00:22, 1056.55it/s]\nTokenizing inputs:  41%|████      | 16383/40168 [00:09<00:22, 1080.90it/s]\nTokenizing inputs:  41%|████      | 16494/40168 [00:09<00:21, 1080.54it/s]\nTokenizing inputs:  41%|████▏     | 16604/40168 [00:10<00:21, 1083.08it/s]\nTokenizing inputs:  42%|████▏     | 16714/40168 [00:10<00:21, 1068.95it/s]\nTokenizing inputs:  42%|████▏     | 16825/40168 [00:10<00:21, 1078.53it/s]\nTokenizing inputs:  42%|████▏     | 16934/40168 [00:10<00:21, 1075.74it/s]\nTokenizing inputs:  42%|████▏     | 17042/40168 [00:10<00:21, 1069.95it/s]\nTokenizing inputs:  43%|████▎     | 17152/40168 [00:10<00:21, 1076.08it/s]\nTokenizing inputs:  43%|████▎     | 17270/40168 [00:10<00:20, 1104.88it/s]\nTokenizing inputs:  43%|████▎     | 17381/40168 [00:10<00:20, 1104.08it/s]\nTokenizing inputs:  44%|████▎     | 17492/40168 [00:10<00:20, 1100.52it/s]\nTokenizing inputs:  44%|████▍     | 17607/40168 [00:10<00:20, 1114.51it/s]\nTokenizing inputs:  44%|████▍     | 17719/40168 [00:11<00:20, 1097.11it/s]\nTokenizing inputs:  44%|████▍     | 17829/40168 [00:11<00:20, 1092.80it/s]\nTokenizing inputs:  45%|████▍     | 17940/40168 [00:11<00:20, 1097.78it/s]\nTokenizing inputs:  45%|████▍     | 18054/40168 [00:11<00:19, 1110.29it/s]\nTokenizing inputs:  45%|████▌     | 18166/40168 [00:11<00:19, 1106.67it/s]\nTokenizing inputs:  46%|████▌     | 18279/40168 [00:11<00:19, 1111.43it/s]\nTokenizing inputs:  46%|████▌     | 18396/40168 [00:11<00:19, 1128.53it/s]\nTokenizing inputs:  46%|████▌     | 18509/40168 [00:11<00:19, 1086.53it/s]\nTokenizing inputs:  46%|████▋     | 18621/40168 [00:11<00:19, 1094.58it/s]\nTokenizing inputs:  47%|████▋     | 18732/40168 [00:11<00:19, 1097.42it/s]\nTokenizing inputs:  47%|████▋     | 18842/40168 [00:12<00:21, 1003.29it/s]\nTokenizing inputs:  47%|████▋     | 18953/40168 [00:12<00:20, 1030.97it/s]\nTokenizing inputs:  47%|████▋     | 19060/40168 [00:12<00:20, 1041.92it/s]\nTokenizing inputs:  48%|████▊     | 19175/40168 [00:12<00:19, 1071.97it/s]\nTokenizing inputs:  48%|████▊     | 19286/40168 [00:12<00:19, 1082.67it/s]\nTokenizing inputs:  48%|████▊     | 19397/40168 [00:12<00:19, 1088.73it/s]\nTokenizing inputs:  49%|████▊     | 19507/40168 [00:12<00:18, 1090.83it/s]\nTokenizing inputs:  49%|████▉     | 19621/40168 [00:12<00:18, 1102.92it/s]\nTokenizing inputs:  49%|████▉     | 19732/40168 [00:12<00:18, 1095.60it/s]\nTokenizing inputs:  49%|████▉     | 19846/40168 [00:12<00:18, 1107.20it/s]\nTokenizing inputs:  50%|████▉     | 19957/40168 [00:13<00:18, 1092.57it/s]\nTokenizing inputs:  50%|████▉     | 20071/40168 [00:13<00:18, 1105.37it/s]\nTokenizing inputs:  50%|█████     | 20182/40168 [00:13<00:18, 1097.71it/s]\nTokenizing inputs:  51%|█████     | 20292/40168 [00:13<00:18, 1093.07it/s]\nTokenizing inputs:  51%|█████     | 20402/40168 [00:13<00:18, 1079.58it/s]\nTokenizing inputs:  51%|█████     | 20513/40168 [00:13<00:18, 1087.38it/s]\nTokenizing inputs:  51%|█████▏    | 20623/40168 [00:13<00:17, 1089.31it/s]\nTokenizing inputs:  52%|█████▏    | 20732/40168 [00:13<00:18, 1076.49it/s]\nTokenizing inputs:  52%|█████▏    | 20847/40168 [00:13<00:17, 1097.46it/s]\nTokenizing inputs:  52%|█████▏    | 20957/40168 [00:14<00:18, 1026.25it/s]\nTokenizing inputs:  52%|█████▏    | 21061/40168 [00:14<00:19, 999.64it/s] \nTokenizing inputs:  53%|█████▎    | 21162/40168 [00:14<00:19, 996.37it/s]\nTokenizing inputs:  53%|█████▎    | 21274/40168 [00:14<00:18, 1030.56it/s]\nTokenizing inputs:  53%|█████▎    | 21383/40168 [00:14<00:17, 1044.44it/s]\nTokenizing inputs:  54%|█████▎    | 21493/40168 [00:14<00:17, 1057.65it/s]\nTokenizing inputs:  54%|█████▍    | 21604/40168 [00:14<00:17, 1071.87it/s]\nTokenizing inputs:  54%|█████▍    | 21713/40168 [00:14<00:17, 1076.80it/s]\nTokenizing inputs:  54%|█████▍    | 21829/40168 [00:14<00:16, 1099.09it/s]\nTokenizing inputs:  55%|█████▍    | 21943/40168 [00:14<00:16, 1108.92it/s]\nTokenizing inputs:  55%|█████▍    | 22054/40168 [00:15<00:16, 1089.30it/s]\nTokenizing inputs:  55%|█████▌    | 22164/40168 [00:15<00:16, 1081.36it/s]\nTokenizing inputs:  55%|█████▌    | 22281/40168 [00:15<00:16, 1103.95it/s]\nTokenizing inputs:  56%|█████▌    | 22397/40168 [00:15<00:15, 1117.32it/s]\nTokenizing inputs:  56%|█████▌    | 22509/40168 [00:15<00:15, 1108.98it/s]\nTokenizing inputs:  56%|█████▋    | 22620/40168 [00:15<00:15, 1108.60it/s]\nTokenizing inputs:  57%|█████▋    | 22736/40168 [00:15<00:15, 1121.51it/s]\nTokenizing inputs:  57%|█████▋    | 22851/40168 [00:15<00:15, 1127.59it/s]\nTokenizing inputs:  57%|█████▋    | 22964/40168 [00:15<00:15, 1127.01it/s]\nTokenizing inputs:  57%|█████▋    | 23077/40168 [00:15<00:15, 1114.76it/s]\nTokenizing inputs:  58%|█████▊    | 23189/40168 [00:16<00:15, 1088.82it/s]\nTokenizing inputs:  58%|█████▊    | 23306/40168 [00:16<00:15, 1109.59it/s]\nTokenizing inputs:  58%|█████▊    | 23421/40168 [00:16<00:14, 1120.65it/s]\nTokenizing inputs:  59%|█████▊    | 23536/40168 [00:16<00:14, 1126.53it/s]\nTokenizing inputs:  59%|█████▉    | 23649/40168 [00:16<00:14, 1105.32it/s]\nTokenizing inputs:  59%|█████▉    | 23761/40168 [00:16<00:14, 1106.55it/s]\nTokenizing inputs:  59%|█████▉    | 23872/40168 [00:16<00:14, 1105.43it/s]\nTokenizing inputs:  60%|█████▉    | 23983/40168 [00:16<00:14, 1099.27it/s]\nTokenizing inputs:  60%|█████▉    | 24093/40168 [00:16<00:14, 1098.98it/s]\nTokenizing inputs:  60%|██████    | 24203/40168 [00:16<00:14, 1093.58it/s]\nTokenizing inputs:  61%|██████    | 24313/40168 [00:17<00:14, 1088.40it/s]\nTokenizing inputs:  61%|██████    | 24423/40168 [00:17<00:14, 1090.06it/s]\nTokenizing inputs:  61%|██████    | 24535/40168 [00:17<00:14, 1097.01it/s]\nTokenizing inputs:  61%|██████▏   | 24645/40168 [00:17<00:14, 1097.45it/s]\nTokenizing inputs:  62%|██████▏   | 24761/40168 [00:17<00:13, 1114.55it/s]\nTokenizing inputs:  62%|██████▏   | 24873/40168 [00:17<00:14, 1090.67it/s]\nTokenizing inputs:  62%|██████▏   | 24983/40168 [00:17<00:13, 1091.82it/s]\nTokenizing inputs:  62%|██████▏   | 25097/40168 [00:17<00:13, 1105.19it/s]\nTokenizing inputs:  63%|██████▎   | 25208/40168 [00:17<00:13, 1105.35it/s]\nTokenizing inputs:  63%|██████▎   | 25319/40168 [00:18<00:13, 1081.91it/s]\nTokenizing inputs:  63%|██████▎   | 25431/40168 [00:18<00:13, 1090.47it/s]\nTokenizing inputs:  64%|██████▎   | 25548/40168 [00:18<00:13, 1112.97it/s]\nTokenizing inputs:  64%|██████▍   | 25660/40168 [00:18<00:13, 1104.35it/s]\nTokenizing inputs:  64%|██████▍   | 25771/40168 [00:18<00:13, 1096.44it/s]\nTokenizing inputs:  64%|██████▍   | 25886/40168 [00:18<00:12, 1111.33it/s]\nTokenizing inputs:  65%|██████▍   | 25998/40168 [00:18<00:12, 1112.38it/s]\nTokenizing inputs:  65%|██████▌   | 26110/40168 [00:18<00:12, 1109.37it/s]\nTokenizing inputs:  65%|██████▌   | 26221/40168 [00:18<00:12, 1097.22it/s]\nTokenizing inputs:  66%|██████▌   | 26331/40168 [00:18<00:12, 1096.03it/s]\nTokenizing inputs:  66%|██████▌   | 26441/40168 [00:19<00:12, 1078.53it/s]\nTokenizing inputs:  66%|██████▌   | 26549/40168 [00:19<00:12, 1070.73it/s]\nTokenizing inputs:  66%|██████▋   | 26657/40168 [00:19<00:12, 1072.14it/s]\nTokenizing inputs:  67%|██████▋   | 26769/40168 [00:19<00:12, 1085.25it/s]\nTokenizing inputs:  67%|██████▋   | 26878/40168 [00:19<00:12, 1058.01it/s]\nTokenizing inputs:  67%|██████▋   | 26984/40168 [00:19<00:12, 1045.07it/s]\nTokenizing inputs:  67%|██████▋   | 27092/40168 [00:19<00:12, 1054.40it/s]\nTokenizing inputs:  68%|██████▊   | 27198/40168 [00:19<00:12, 1043.01it/s]\nTokenizing inputs:  68%|██████▊   | 27312/40168 [00:19<00:12, 1071.08it/s]\nTokenizing inputs:  68%|██████▊   | 27420/40168 [00:19<00:12, 1061.34it/s]\nTokenizing inputs:  69%|██████▊   | 27532/40168 [00:20<00:11, 1078.09it/s]\nTokenizing inputs:  69%|██████▉   | 27640/40168 [00:20<00:11, 1077.67it/s]\nTokenizing inputs:  69%|██████▉   | 27750/40168 [00:20<00:11, 1083.99it/s]\nTokenizing inputs:  69%|██████▉   | 27859/40168 [00:20<00:11, 1079.67it/s]\nTokenizing inputs:  70%|██████▉   | 27970/40168 [00:20<00:11, 1088.42it/s]\nTokenizing inputs:  70%|██████▉   | 28079/40168 [00:20<00:11, 1085.07it/s]\nTokenizing inputs:  70%|███████   | 28192/40168 [00:20<00:10, 1097.14it/s]\nTokenizing inputs:  70%|███████   | 28302/40168 [00:20<00:10, 1091.01it/s]\nTokenizing inputs:  71%|███████   | 28412/40168 [00:20<00:10, 1087.10it/s]\nTokenizing inputs:  71%|███████   | 28523/40168 [00:20<00:10, 1093.22it/s]\nTokenizing inputs:  71%|███████▏  | 28633/40168 [00:21<00:10, 1075.01it/s]\nTokenizing inputs:  72%|███████▏  | 28745/40168 [00:21<00:10, 1086.98it/s]\nTokenizing inputs:  72%|███████▏  | 28860/40168 [00:21<00:10, 1105.31it/s]\nTokenizing inputs:  72%|███████▏  | 28971/40168 [00:21<00:10, 1095.09it/s]\nTokenizing inputs:  72%|███████▏  | 29081/40168 [00:21<00:10, 1095.82it/s]\nTokenizing inputs:  73%|███████▎  | 29196/40168 [00:21<00:09, 1111.80it/s]\nTokenizing inputs:  73%|███████▎  | 29308/40168 [00:21<00:09, 1089.84it/s]\nTokenizing inputs:  73%|███████▎  | 29418/40168 [00:21<00:09, 1078.47it/s]\nTokenizing inputs:  74%|███████▎  | 29526/40168 [00:21<00:09, 1074.51it/s]\nTokenizing inputs:  74%|███████▍  | 29634/40168 [00:21<00:09, 1075.85it/s]\nTokenizing inputs:  74%|███████▍  | 29742/40168 [00:22<00:10, 978.85it/s] \nTokenizing inputs:  74%|███████▍  | 29851/40168 [00:22<00:10, 1008.19it/s]\nTokenizing inputs:  75%|███████▍  | 29967/40168 [00:22<00:09, 1048.53it/s]\nTokenizing inputs:  75%|███████▍  | 30077/40168 [00:22<00:09, 1060.35it/s]\nTokenizing inputs:  75%|███████▌  | 30190/40168 [00:22<00:09, 1080.18it/s]\nTokenizing inputs:  75%|███████▌  | 30300/40168 [00:22<00:09, 1084.05it/s]\nTokenizing inputs:  76%|███████▌  | 30410/40168 [00:22<00:08, 1085.66it/s]\nTokenizing inputs:  76%|███████▌  | 30519/40168 [00:22<00:08, 1073.44it/s]\nTokenizing inputs:  76%|███████▌  | 30627/40168 [00:22<00:08, 1065.65it/s]\nTokenizing inputs:  77%|███████▋  | 30738/40168 [00:23<00:08, 1076.41it/s]\nTokenizing inputs:  77%|███████▋  | 30846/40168 [00:23<00:08, 1059.29it/s]\nTokenizing inputs:  77%|███████▋  | 30960/40168 [00:23<00:08, 1080.87it/s]\nTokenizing inputs:  77%|███████▋  | 31075/40168 [00:23<00:08, 1099.89it/s]\nTokenizing inputs:  78%|███████▊  | 31186/40168 [00:23<00:08, 1096.07it/s]\nTokenizing inputs:  78%|███████▊  | 31296/40168 [00:23<00:08, 1089.23it/s]\nTokenizing inputs:  78%|███████▊  | 31405/40168 [00:23<00:08, 1085.48it/s]\nTokenizing inputs:  78%|███████▊  | 31514/40168 [00:23<00:07, 1086.01it/s]\nTokenizing inputs:  79%|███████▊  | 31623/40168 [00:23<00:07, 1071.75it/s]\nTokenizing inputs:  79%|███████▉  | 31732/40168 [00:23<00:07, 1075.99it/s]\nTokenizing inputs:  79%|███████▉  | 31840/40168 [00:24<00:07, 1061.85it/s]\nTokenizing inputs:  80%|███████▉  | 31953/40168 [00:24<00:07, 1080.19it/s]\nTokenizing inputs:  80%|███████▉  | 32064/40168 [00:24<00:07, 1086.50it/s]\nTokenizing inputs:  80%|████████  | 32178/40168 [00:24<00:07, 1100.37it/s]\nTokenizing inputs:  80%|████████  | 32293/40168 [00:24<00:07, 1112.86it/s]\nTokenizing inputs:  81%|████████  | 32407/40168 [00:24<00:06, 1117.71it/s]\nTokenizing inputs:  81%|████████  | 32521/40168 [00:24<00:06, 1122.49it/s]\nTokenizing inputs:  81%|████████▏ | 32640/40168 [00:24<00:06, 1140.08it/s]\nTokenizing inputs:  82%|████████▏ | 32755/40168 [00:24<00:06, 1112.95it/s]\nTokenizing inputs:  82%|████████▏ | 32867/40168 [00:24<00:06, 1094.49it/s]\nTokenizing inputs:  82%|████████▏ | 32977/40168 [00:25<00:06, 1077.69it/s]\nTokenizing inputs:  82%|████████▏ | 33085/40168 [00:25<00:06, 1073.74it/s]\nTokenizing inputs:  83%|████████▎ | 33197/40168 [00:25<00:06, 1085.57it/s]\nTokenizing inputs:  83%|████████▎ | 33306/40168 [00:25<00:06, 1079.15it/s]\nTokenizing inputs:  83%|████████▎ | 33414/40168 [00:25<00:06, 1068.85it/s]\nTokenizing inputs:  83%|████████▎ | 33521/40168 [00:25<00:06, 1046.79it/s]\nTokenizing inputs:  84%|████████▎ | 33626/40168 [00:25<00:06, 1045.97it/s]\nTokenizing inputs:  84%|████████▍ | 33731/40168 [00:25<00:06, 1046.24it/s]\nTokenizing inputs:  84%|████████▍ | 33838/40168 [00:25<00:06, 1053.08it/s]\nTokenizing inputs:  85%|████████▍ | 33944/40168 [00:26<00:05, 1043.71it/s]\nTokenizing inputs:  85%|████████▍ | 34049/40168 [00:26<00:05, 1033.96it/s]\nTokenizing inputs:  85%|████████▌ | 34162/40168 [00:26<00:05, 1060.60it/s]\nTokenizing inputs:  85%|████████▌ | 34270/40168 [00:26<00:05, 1063.40it/s]\nTokenizing inputs:  86%|████████▌ | 34381/40168 [00:26<00:05, 1075.66it/s]\nTokenizing inputs:  86%|████████▌ | 34489/40168 [00:26<00:05, 1064.66it/s]\nTokenizing inputs:  86%|████████▌ | 34598/40168 [00:26<00:05, 1069.79it/s]\nTokenizing inputs:  86%|████████▋ | 34711/40168 [00:26<00:05, 1086.29it/s]\nTokenizing inputs:  87%|████████▋ | 34820/40168 [00:26<00:04, 1084.95it/s]\nTokenizing inputs:  87%|████████▋ | 34929/40168 [00:26<00:04, 1068.71it/s]\nTokenizing inputs:  87%|████████▋ | 35037/40168 [00:27<00:04, 1069.76it/s]\nTokenizing inputs:  87%|████████▋ | 35145/40168 [00:27<00:04, 1056.61it/s]\nTokenizing inputs:  88%|████████▊ | 35256/40168 [00:27<00:04, 1071.99it/s]\nTokenizing inputs:  88%|████████▊ | 35364/40168 [00:27<00:04, 1057.49it/s]\nTokenizing inputs:  88%|████████▊ | 35470/40168 [00:27<00:04, 1041.69it/s]\nTokenizing inputs:  89%|████████▊ | 35575/40168 [00:27<00:04, 1025.27it/s]\nTokenizing inputs:  89%|████████▉ | 35681/40168 [00:27<00:04, 1034.51it/s]\nTokenizing inputs:  89%|████████▉ | 35787/40168 [00:27<00:04, 1040.12it/s]\nTokenizing inputs:  89%|████████▉ | 35893/40168 [00:27<00:04, 1043.17it/s]\nTokenizing inputs:  90%|████████▉ | 36000/40168 [00:27<00:03, 1048.50it/s]\nTokenizing inputs:  90%|████████▉ | 36105/40168 [00:28<00:03, 1029.98it/s]\nTokenizing inputs:  90%|█████████ | 36209/40168 [00:28<00:03, 1027.16it/s]\nTokenizing inputs:  90%|█████████ | 36313/40168 [00:28<00:03, 1029.93it/s]\nTokenizing inputs:  91%|█████████ | 36420/40168 [00:28<00:03, 1039.42it/s]\nTokenizing inputs:  91%|█████████ | 36524/40168 [00:28<00:03, 1019.08it/s]\nTokenizing inputs:  91%|█████████ | 36627/40168 [00:28<00:03, 1021.63it/s]\nTokenizing inputs:  91%|█████████▏| 36730/40168 [00:28<00:03, 1022.51it/s]\nTokenizing inputs:  92%|█████████▏| 36834/40168 [00:28<00:03, 1025.80it/s]\nTokenizing inputs:  92%|█████████▏| 36939/40168 [00:28<00:03, 1030.37it/s]\nTokenizing inputs:  92%|█████████▏| 37049/40168 [00:28<00:02, 1047.33it/s]\nTokenizing inputs:  92%|█████████▏| 37154/40168 [00:29<00:02, 1019.40it/s]\nTokenizing inputs:  93%|█████████▎| 37257/40168 [00:29<00:02, 1019.85it/s]\nTokenizing inputs:  93%|█████████▎| 37360/40168 [00:29<00:02, 1012.40it/s]\nTokenizing inputs:  93%|█████████▎| 37465/40168 [00:29<00:02, 1022.74it/s]\nTokenizing inputs:  94%|█████████▎| 37572/40168 [00:29<00:02, 1035.16it/s]\nTokenizing inputs:  94%|█████████▍| 37676/40168 [00:29<00:02, 1013.52it/s]\nTokenizing inputs:  94%|█████████▍| 37780/40168 [00:29<00:02, 1018.88it/s]\nTokenizing inputs:  94%|█████████▍| 37882/40168 [00:29<00:02, 1015.97it/s]\nTokenizing inputs:  95%|█████████▍| 37989/40168 [00:29<00:02, 1030.20it/s]\nTokenizing inputs:  95%|█████████▍| 38100/40168 [00:29<00:01, 1053.49it/s]\nTokenizing inputs:  95%|█████████▌| 38206/40168 [00:30<00:01, 1042.02it/s]\nTokenizing inputs:  95%|█████████▌| 38311/40168 [00:30<00:01, 1030.79it/s]\nTokenizing inputs:  96%|█████████▌| 38415/40168 [00:30<00:01, 1024.18it/s]\nTokenizing inputs:  96%|█████████▌| 38523/40168 [00:30<00:01, 1037.86it/s]\nTokenizing inputs:  96%|█████████▌| 38630/40168 [00:30<00:01, 1046.08it/s]\nTokenizing inputs:  96%|█████████▋| 38735/40168 [00:30<00:01, 1035.89it/s]\nTokenizing inputs:  97%|█████████▋| 38842/40168 [00:30<00:01, 1044.79it/s]\nTokenizing inputs:  97%|█████████▋| 38953/40168 [00:30<00:01, 1061.48it/s]\nTokenizing inputs:  97%|█████████▋| 39060/40168 [00:30<00:01, 1062.88it/s]\nTokenizing inputs:  98%|█████████▊| 39167/40168 [00:31<00:00, 1060.52it/s]\nTokenizing inputs:  98%|█████████▊| 39274/40168 [00:31<00:00, 1049.20it/s]\nTokenizing inputs:  98%|█████████▊| 39381/40168 [00:31<00:00, 1055.06it/s]\nTokenizing inputs:  98%|█████████▊| 39487/40168 [00:31<00:00, 1052.82it/s]\nTokenizing inputs:  99%|█████████▊| 39594/40168 [00:31<00:00, 1054.77it/s]\nTokenizing inputs:  99%|█████████▉| 39700/40168 [00:31<00:00, 1031.38it/s]\nTokenizing inputs:  99%|█████████▉| 39811/40168 [00:31<00:00, 1052.42it/s]\nTokenizing inputs:  99%|█████████▉| 39917/40168 [00:31<00:00, 1034.60it/s]\nTokenizing inputs: 100%|█████████▉| 40029/40168 [00:31<00:00, 1055.02it/s]\nTokenizing inputs: 100%|█████████▉| 40135/40168 [00:31<00:00, 1049.15it/s]\nTokenizing inputs: 100%|██████████| 40168/40168 [00:31<00:00, 1256.47it/s]\n\nRunning loglikelihood requests:   0%|          | 0/40168 [00:00<?, ?it/s]\nRunning loglikelihood requests:   0%|          | 1/40168 [00:10<116:59:56, 10.49s/it]\nRunning loglikelihood requests:   1%|          | 469/40168 [00:10<10:30, 62.99it/s]  \nRunning loglikelihood requests:   2%|▏         | 937/40168 [00:18<10:47, 60.58it/s]\nRunning loglikelihood requests:   3%|▎         | 1134/40168 [00:25<14:21, 45.32it/s]\nRunning loglikelihood requests:   4%|▍         | 1538/40168 [00:33<13:13, 48.66it/s]\nRunning loglikelihood requests:   5%|▍         | 2007/40168 [00:33<07:39, 83.02it/s]\nRunning loglikelihood requests:   6%|▌         | 2220/40168 [00:40<10:39, 59.33it/s]\nRunning loglikelihood requests:   6%|▋         | 2562/40168 [00:48<11:25, 54.83it/s]\nRunning loglikelihood requests:   8%|▊         | 3031/40168 [00:48<06:55, 89.46it/s]\nRunning loglikelihood requests:   8%|▊         | 3246/40168 [00:55<09:43, 63.26it/s]\nRunning loglikelihood requests:   9%|▉         | 3586/40168 [01:02<10:43, 56.88it/s]\nRunning loglikelihood requests:  10%|█         | 4045/40168 [01:02<06:37, 90.81it/s]\nRunning loglikelihood requests:  11%|█         | 4258/40168 [01:10<09:20, 64.12it/s]\nRunning loglikelihood requests:  11%|█▏        | 4611/40168 [01:17<10:07, 58.53it/s]\nRunning loglikelihood requests:  13%|█▎        | 5067/40168 [01:17<06:19, 92.59it/s]\nRunning loglikelihood requests:  13%|█▎        | 5279/40168 [01:24<08:44, 66.50it/s]\nRunning loglikelihood requests:  14%|█▍        | 5636/40168 [01:31<09:28, 60.73it/s]\nRunning loglikelihood requests:  15%|█▌        | 6079/40168 [01:31<05:59, 94.77it/s]\nRunning loglikelihood requests:  16%|█▌        | 6286/40168 [01:38<08:30, 66.36it/s]\nRunning loglikelihood requests:  17%|█▋        | 6662/40168 [01:45<09:05, 61.44it/s]\nRunning loglikelihood requests:  18%|█▊        | 7125/40168 [01:45<05:40, 97.06it/s]\nRunning loglikelihood requests:  18%|█▊        | 7337/40168 [01:52<08:02, 68.03it/s]\nRunning loglikelihood requests:  19%|█▉        | 7687/40168 [01:59<08:48, 61.50it/s]\nRunning loglikelihood requests:  20%|██        | 8138/40168 [01:59<05:32, 96.48it/s]\nRunning loglikelihood requests:  21%|██        | 8348/40168 [02:06<07:47, 68.02it/s]\nRunning loglikelihood requests:  22%|██▏       | 8713/40168 [02:12<08:21, 62.67it/s]\nRunning loglikelihood requests:  23%|██▎       | 9190/40168 [02:12<05:09, 100.23it/s]\nRunning loglikelihood requests:  23%|██▎       | 9407/40168 [02:19<07:17, 70.37it/s] \nRunning loglikelihood requests:  24%|██▍       | 9740/40168 [02:26<08:05, 62.68it/s]\nRunning loglikelihood requests:  25%|██▌       | 10203/40168 [02:26<05:01, 99.55it/s]\nRunning loglikelihood requests:  26%|██▌       | 10418/40168 [02:33<07:02, 70.39it/s]\nRunning loglikelihood requests:  27%|██▋       | 10765/40168 [02:39<07:39, 64.00it/s]\nRunning loglikelihood requests:  28%|██▊       | 11094/40168 [02:39<05:21, 90.55it/s]\nRunning loglikelihood requests:  28%|██▊       | 11277/40168 [02:46<07:33, 63.66it/s]\nRunning loglikelihood requests:  29%|██▉       | 11725/40168 [02:46<04:29, 105.39it/s]\nRunning loglikelihood requests:  30%|██▉       | 11946/40168 [02:52<06:30, 72.31it/s] \nRunning loglikelihood requests:  31%|███       | 12301/40168 [02:59<07:06, 65.35it/s]\nRunning loglikelihood requests:  32%|███▏      | 12763/40168 [02:59<04:20, 105.26it/s]\nRunning loglikelihood requests:  32%|███▏      | 12979/40168 [03:05<06:03, 74.77it/s] \nRunning loglikelihood requests:  33%|███▎      | 13326/40168 [03:11<06:36, 67.62it/s]\nRunning loglikelihood requests:  34%|███▍      | 13783/40168 [03:11<04:05, 107.29it/s]\nRunning loglikelihood requests:  35%|███▍      | 13996/40168 [03:17<05:45, 75.86it/s] \nRunning loglikelihood requests:  36%|███▌      | 14351/40168 [03:23<06:12, 69.22it/s]\nRunning loglikelihood requests:  37%|███▋      | 14833/40168 [03:23<03:47, 111.52it/s]\nRunning loglikelihood requests:  37%|███▋      | 15052/40168 [03:29<05:17, 79.12it/s] \nRunning loglikelihood requests:  38%|███▊      | 15375/40168 [03:35<05:56, 69.53it/s]\nRunning loglikelihood requests:  39%|███▉      | 15841/40168 [03:36<03:39, 110.98it/s]\nRunning loglikelihood requests:  40%|███▉      | 16058/40168 [03:41<05:05, 78.87it/s] \nRunning loglikelihood requests:  41%|████      | 16400/40168 [03:47<05:34, 70.97it/s]\nRunning loglikelihood requests:  42%|████▏     | 16880/40168 [03:47<03:23, 114.31it/s]\nRunning loglikelihood requests:  43%|████▎     | 17100/40168 [03:53<04:44, 81.18it/s] \nRunning loglikelihood requests:  43%|████▎     | 17426/40168 [03:59<05:16, 71.75it/s]\nRunning loglikelihood requests:  44%|████▍     | 17829/40168 [03:59<03:25, 108.58it/s]\nRunning loglikelihood requests:  45%|████▍     | 18027/40168 [04:05<04:48, 76.65it/s] \nRunning loglikelihood requests:  46%|████▌     | 18451/40168 [04:11<04:46, 75.82it/s]\nRunning loglikelihood requests:  47%|████▋     | 18906/40168 [04:11<02:59, 118.13it/s]\nRunning loglikelihood requests:  48%|████▊     | 19113/40168 [04:16<04:13, 83.13it/s] \nRunning loglikelihood requests:  48%|████▊     | 19475/40168 [04:22<04:30, 76.37it/s]\nRunning loglikelihood requests:  50%|████▉     | 19958/40168 [04:22<02:45, 122.27it/s]\nRunning loglikelihood requests:  50%|█████     | 20176/40168 [04:27<03:50, 86.88it/s] \nRunning loglikelihood requests:  51%|█████     | 20501/40168 [04:33<04:17, 76.32it/s]\nRunning loglikelihood requests:  52%|█████▏    | 20976/40168 [04:33<02:36, 122.26it/s]\nRunning loglikelihood requests:  53%|█████▎    | 21195/40168 [04:38<03:38, 86.95it/s] \nRunning loglikelihood requests:  54%|█████▎    | 21525/40168 [04:44<03:59, 77.99it/s]\nRunning loglikelihood requests:  55%|█████▍    | 21989/40168 [04:44<02:26, 124.04it/s]\nRunning loglikelihood requests:  55%|█████▌    | 22205/40168 [04:49<03:20, 89.54it/s] \nRunning loglikelihood requests:  56%|█████▌    | 22549/40168 [04:54<03:37, 81.13it/s]\nRunning loglikelihood requests:  57%|█████▋    | 22978/40168 [04:54<02:17, 125.12it/s]\nRunning loglikelihood requests:  58%|█████▊    | 23183/40168 [04:59<03:10, 89.30it/s] \nRunning loglikelihood requests:  59%|█████▊    | 23574/40168 [05:04<03:14, 85.36it/s]\nRunning loglikelihood requests:  60%|█████▉    | 24047/40168 [05:04<01:59, 135.33it/s]\nRunning loglikelihood requests:  60%|██████    | 24262/40168 [05:09<02:45, 96.31it/s] \nRunning loglikelihood requests:  61%|██████    | 24599/40168 [05:14<02:59, 86.51it/s]\nRunning loglikelihood requests:  62%|██████▏   | 25071/40168 [05:14<01:49, 137.82it/s]\nRunning loglikelihood requests:  63%|██████▎   | 25288/40168 [05:18<02:30, 98.89it/s] \nRunning loglikelihood requests:  64%|██████▍   | 25624/40168 [05:23<02:42, 89.33it/s]\nRunning loglikelihood requests:  65%|██████▍   | 26075/40168 [05:23<01:40, 140.12it/s]\nRunning loglikelihood requests:  65%|██████▌   | 26287/40168 [05:28<02:17, 100.74it/s]\nRunning loglikelihood requests:  66%|██████▋   | 26648/40168 [05:32<02:22, 95.01it/s] \nRunning loglikelihood requests:  68%|██████▊   | 27138/40168 [05:32<01:25, 153.21it/s]\nRunning loglikelihood requests:  68%|██████▊   | 27360/40168 [05:36<01:53, 112.98it/s]\nRunning loglikelihood requests:  69%|██████▉   | 27672/40168 [05:40<02:03, 101.30it/s]\nRunning loglikelihood requests:  70%|███████   | 28176/40168 [05:40<01:12, 166.18it/s]\nRunning loglikelihood requests:  71%|███████   | 28406/40168 [05:44<01:35, 123.37it/s]\nRunning loglikelihood requests:  71%|███████▏  | 28696/40168 [05:47<01:45, 108.65it/s]\nRunning loglikelihood requests:  73%|███████▎  | 29188/40168 [05:47<01:01, 177.39it/s]\nRunning loglikelihood requests:  73%|███████▎  | 29418/40168 [05:51<01:20, 133.82it/s]\nRunning loglikelihood requests:  74%|███████▍  | 29720/40168 [05:54<01:26, 120.19it/s]\nRunning loglikelihood requests:  75%|███████▌  | 30216/40168 [05:54<00:50, 196.52it/s]\nRunning loglikelihood requests:  76%|███████▌  | 30447/40168 [05:57<01:05, 148.05it/s]\nRunning loglikelihood requests:  77%|███████▋  | 30744/40168 [06:00<01:11, 130.96it/s]\nRunning loglikelihood requests:  78%|███████▊  | 31244/40168 [06:00<00:41, 214.69it/s]\nRunning loglikelihood requests:  78%|███████▊  | 31477/40168 [06:03<00:54, 160.93it/s]\nRunning loglikelihood requests:  79%|███████▉  | 31768/40168 [06:06<00:59, 140.38it/s]\nRunning loglikelihood requests:  80%|████████  | 32264/40168 [06:06<00:34, 229.38it/s]\nRunning loglikelihood requests:  81%|████████  | 32496/40168 [06:08<00:44, 171.52it/s]\nRunning loglikelihood requests:  82%|████████▏ | 32792/40168 [06:11<00:48, 151.61it/s]\nRunning loglikelihood requests:  83%|████████▎ | 33304/40168 [06:13<00:40, 168.87it/s]\nRunning loglikelihood requests:  84%|████████▍ | 33813/40168 [06:13<00:24, 261.39it/s]\nRunning loglikelihood requests:  85%|████████▍ | 34028/40168 [06:16<00:31, 194.91it/s]\nRunning loglikelihood requests:  85%|████████▌ | 34328/40168 [06:18<00:33, 172.30it/s]\nRunning loglikelihood requests:  87%|████████▋ | 34825/40168 [06:18<00:19, 274.42it/s]\nRunning loglikelihood requests:  87%|████████▋ | 35052/40168 [06:20<00:24, 205.15it/s]\nRunning loglikelihood requests:  88%|████████▊ | 35352/40168 [06:23<00:26, 181.13it/s]\nRunning loglikelihood requests:  89%|████████▉ | 35859/40168 [06:23<00:14, 294.42it/s]\nRunning loglikelihood requests:  90%|████████▉ | 36093/40168 [06:25<00:18, 220.85it/s]\nRunning loglikelihood requests:  91%|█████████ | 36376/40168 [06:27<00:19, 191.07it/s]\nRunning loglikelihood requests:  92%|█████████▏| 36869/40168 [06:27<00:10, 309.02it/s]\nRunning loglikelihood requests:  92%|█████████▏| 37102/40168 [06:29<00:13, 232.47it/s]\nRunning loglikelihood requests:  93%|█████████▎| 37400/40168 [06:31<00:13, 206.86it/s]\nRunning loglikelihood requests:  94%|█████████▍| 37897/40168 [06:31<00:06, 334.93it/s]\nRunning loglikelihood requests:  95%|█████████▍| 38131/40168 [06:33<00:08, 253.43it/s]\nRunning loglikelihood requests:  96%|█████████▌| 38424/40168 [06:34<00:07, 228.48it/s]\nRunning loglikelihood requests:  97%|█████████▋| 38921/40168 [06:34<00:03, 369.19it/s]\nRunning loglikelihood requests:  97%|█████████▋| 39156/40168 [06:36<00:03, 288.32it/s]\nRunning loglikelihood requests:  98%|█████████▊| 39448/40168 [06:37<00:02, 265.54it/s]\nRunning loglikelihood requests:  99%|█████████▉| 39960/40168 [06:38<00:00, 373.86it/s]\nRunning loglikelihood requests: 100%|██████████| 40168/40168 [06:38<00:00, 100.88it/s]\nfatal: not a git repository (or any parent up to mount point /kaggle)\nStopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).\n{\n  \"hellaswag\": {\n    \"name\": \"hellaswag\",\n    \"alias\": \"hellaswag\",\n    \"sample_len\": 10042,\n    \"acc,none\": 0.3270264887472615,\n    \"acc_stderr,none\": 0.004681682605348133,\n    \"acc_norm,none\": 0.3867755427205736,\n    \"acc_norm_stderr,none\": 0.004860162076330819\n  }\n}\nVoid eval metadata: {\n  \"max_length\": 2048,\n  \"long_continuation_policy\": \"sliding_disjoint_segments\",\n  \"long_continuation_stride\": 256,\n  \"long_request_count\": 0,\n  \"long_continuation_tokens\": 0,\n  \"max_continuation_tokens\": 0\n}\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"  model           model_id                                  revision  \\\n0  Void  appvoid/void-byte  92c601560cc2179e3bfc33e067fdd3bf9254e163   \n\n   hellaswag  \n0  38.677554  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>model</th>\n      <th>model_id</th>\n      <th>revision</th>\n      <th>hellaswag</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Void</td>\n      <td>appvoid/void-byte</td>\n      <td>92c601560cc2179e3bfc33e067fdd3bf9254e163</td>\n      <td>38.677554</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"\n================================================================================\nVOID — piqa\n================================================================================\n\n[piqa] trying fixed batch=512\n\nCOMMAND: /usr/bin/python3 /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_run_void_lm_eval.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --revision 92c601560cc2179e3bfc33e067fdd3bf9254e163 --task piqa --device cuda:0 --dtype float16 --batch-size 512 --max-length 2048 --long-cont-stride 256 --out /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/open_slm/piqa/results.json\ntask = piqa batch = 512 max_length = 2048 long_cont_stride = 256\n\nGenerating train split:   0%|          | 0/16113 [00:00<?, ? examples/s]\nGenerating train split: 100%|██████████| 16113/16113 [00:00<00:00, 603622.84 examples/s]\n\nGenerating validation split:   0%|          | 0/1838 [00:00<?, ? examples/s]\nGenerating validation split: 100%|██████████| 1838/1838 [00:00<00:00, 309193.87 examples/s]\n\nGenerating test split:   0%|          | 0/3084 [00:00<?, ? examples/s]\nGenerating test split: 100%|██████████| 3084/3084 [00:00<00:00, 534734.75 examples/s]\nWARNING:lm_eval.evaluator:Overwriting default num_fewshot of piqa from None to 0\n\n  0%|          | 0/1838 [00:00<?, ?it/s]\n  3%|▎         | 64/1838 [00:00<00:02, 636.31it/s]\n  7%|▋         | 128/1838 [00:00<00:02, 636.96it/s]\n 11%|█         | 194/1838 [00:00<00:02, 642.89it/s]\n 14%|█▍        | 260/1838 [00:00<00:02, 649.56it/s]\n 18%|█▊        | 327/1838 [00:00<00:02, 654.78it/s]\n 21%|██▏       | 393/1838 [00:00<00:02, 651.20it/s]\n 25%|██▍       | 459/1838 [00:00<00:02, 652.93it/s]\n 29%|██▊       | 525/1838 [00:00<00:02, 649.03it/s]\n 32%|███▏      | 592/1838 [00:00<00:01, 653.27it/s]\n 36%|███▌      | 659/1838 [00:01<00:01, 657.40it/s]\n 39%|███▉      | 725/1838 [00:01<00:01, 652.73it/s]\n 43%|████▎     | 791/1838 [00:01<00:01, 651.71it/s]\n 47%|████▋     | 857/1838 [00:01<00:01, 648.41it/s]\n 50%|█████     | 924/1838 [00:01<00:01, 652.47it/s]\n 54%|█████▍    | 990/1838 [00:01<00:01, 649.99it/s]\n 57%|█████▋    | 1056/1838 [00:01<00:01, 647.57it/s]\n 61%|██████    | 1121/1838 [00:01<00:01, 644.02it/s]\n 65%|██████▍   | 1187/1838 [00:01<00:01, 647.97it/s]\n 68%|██████▊   | 1252/1838 [00:01<00:00, 648.46it/s]\n 72%|███████▏  | 1318/1838 [00:02<00:00, 651.64it/s]\n 75%|███████▌  | 1384/1838 [00:02<00:00, 649.30it/s]\n 79%|███████▉  | 1449/1838 [00:02<00:00, 644.56it/s]\n 82%|████████▏ | 1514/1838 [00:02<00:00, 644.52it/s]\n 86%|████████▌ | 1584/1838 [00:02<00:00, 659.33it/s]\n 90%|████████▉ | 1652/1838 [00:02<00:00, 665.36it/s]\n 94%|█████████▍| 1724/1838 [00:02<00:00, 678.96it/s]\n 98%|█████████▊| 1796/1838 [00:02<00:00, 690.27it/s]\n100%|██████████| 1838/1838 [00:02<00:00, 657.38it/s]\n\nTokenizing inputs:   0%|          | 0/3676 [00:00<?, ?it/s]\nTokenizing inputs:   8%|▊         | 286/3676 [00:00<00:01, 2845.15it/s]\nTokenizing inputs:  16%|█▌        | 571/3676 [00:00<00:01, 2793.90it/s]\nTokenizing inputs:  23%|██▎       | 851/3676 [00:00<00:01, 2752.98it/s]\nTokenizing inputs:  31%|███       | 1143/3676 [00:00<00:00, 2816.93it/s]\nTokenizing inputs:  39%|███▉      | 1437/3676 [00:00<00:00, 2859.03it/s]\nTokenizing inputs:  47%|████▋     | 1739/3676 [00:00<00:00, 2911.80it/s]\nTokenizing inputs:  55%|█████▌    | 2031/3676 [00:00<00:00, 2874.35it/s]\nTokenizing inputs:  64%|██████▎   | 2335/3676 [00:00<00:00, 2925.07it/s]\nTokenizing inputs:  72%|███████▏  | 2635/3676 [00:00<00:00, 2946.68it/s]\nTokenizing inputs:  80%|███████▉  | 2930/3676 [00:01<00:00, 2915.57it/s]\nTokenizing inputs:  88%|████████▊ | 3245/3676 [00:01<00:00, 2985.58it/s]\nTokenizing inputs:  96%|█████████▋| 3544/3676 [00:01<00:00, 2977.73it/s]\nTokenizing inputs: 100%|██████████| 3676/3676 [00:01<00:00, 2920.62it/s]\n\nRunning loglikelihood requests:   0%|          | 0/3676 [00:00<?, ?it/s]\nRunning loglikelihood requests:   0%|          | 1/3676 [00:19<19:44:48, 19.34s/it]\nRunning loglikelihood requests:  12%|█▏        | 454/3676 [00:19<01:36, 33.26it/s] \nRunning loglikelihood requests:  25%|██▍       | 907/3676 [00:22<00:47, 58.87it/s]\nRunning loglikelihood requests:  30%|██▉       | 1101/3676 [00:25<00:41, 62.74it/s]\nRunning loglikelihood requests:  42%|████▏     | 1537/3676 [00:27<00:23, 92.65it/s]\nRunning loglikelihood requests:  56%|█████▌    | 2043/3676 [00:27<00:10, 160.94it/s]\nRunning loglikelihood requests:  62%|██████▏   | 2273/3676 [00:29<00:08, 156.25it/s]\nRunning loglikelihood requests:  70%|██████▉   | 2561/3676 [00:30<00:06, 170.12it/s]\nRunning loglikelihood requests:  84%|████████▎ | 3073/3676 [00:31<00:02, 233.12it/s]\nRunning loglikelihood requests:  98%|█████████▊| 3585/3676 [00:31<00:00, 352.79it/s]\nRunning loglikelihood requests: 100%|██████████| 3676/3676 [00:31<00:00, 115.27it/s]\nfatal: not a git repository (or any parent up to mount point /kaggle)\nStopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).\n{\n  \"piqa\": {\n    \"name\": \"piqa\",\n    \"alias\": \"piqa\",\n    \"sample_len\": 1838,\n    \"acc,none\": 0.6599564744287268,\n    \"acc_stderr,none\": 0.011052749414423359,\n    \"acc_norm,none\": 0.6746463547334058,\n    \"acc_norm_stderr,none\": 0.010931036623525334\n  }\n}\nVoid eval metadata: {\n  \"max_length\": 2048,\n  \"long_continuation_policy\": \"sliding_disjoint_segments\",\n  \"long_continuation_stride\": 256,\n  \"long_request_count\": 0,\n  \"long_continuation_tokens\": 0,\n  \"max_continuation_tokens\": 0\n}\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"  model           model_id                                  revision  \\\n0  Void  appvoid/void-byte  92c601560cc2179e3bfc33e067fdd3bf9254e163   \n\n   hellaswag       piqa  \n0  38.677554  67.464635  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>model</th>\n      <th>model_id</th>\n      <th>revision</th>\n      <th>hellaswag</th>\n      <th>piqa</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Void</td>\n      <td>appvoid/void-byte</td>\n      <td>92c601560cc2179e3bfc33e067fdd3bf9254e163</td>\n      <td>38.677554</td>\n      <td>67.464635</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"\n================================================================================\nVOID — arc_easy\n================================================================================\n\n[arc_easy] trying fixed batch=512\n\nCOMMAND: /usr/bin/python3 /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_run_void_lm_eval.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --revision 92c601560cc2179e3bfc33e067fdd3bf9254e163 --task arc_easy --device cuda:0 --dtype float16 --batch-size 512 --max-length 2048 --long-cont-stride 256 --out /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/open_slm/arc_easy/results.json\ntask = arc_easy batch = 512 max_length = 2048 long_cont_stride = 256\n\nGenerating train split:   0%|          | 0/2251 [00:00<?, ? examples/s]\nGenerating train split: 100%|██████████| 2251/2251 [00:00<00:00, 263887.82 examples/s]\n\nGenerating test split:   0%|          | 0/2376 [00:00<?, ? examples/s]\nGenerating test split: 100%|██████████| 2376/2376 [00:00<00:00, 349451.80 examples/s]\n\nGenerating validation split:   0%|          | 0/570 [00:00<?, ? examples/s]\nGenerating validation split: 100%|██████████| 570/570 [00:00<00:00, 138345.77 examples/s]\nWARNING:lm_eval.evaluator:Overwriting default num_fewshot of arc_easy from None to 0\n\n  0%|          | 0/2376 [00:00<?, ?it/s]\n  2%|▏         | 51/2376 [00:00<00:04, 501.20it/s]\n  4%|▍         | 102/2376 [00:00<00:04, 498.77it/s]\n  6%|▋         | 153/2376 [00:00<00:04, 501.59it/s]\n  9%|▊         | 204/2376 [00:00<00:04, 498.72it/s]\n 11%|█         | 254/2376 [00:00<00:04, 496.03it/s]\n 13%|█▎        | 305/2376 [00:00<00:04, 500.24it/s]\n 15%|█▍        | 356/2376 [00:00<00:04, 502.15it/s]\n 17%|█▋        | 407/2376 [00:00<00:03, 502.65it/s]\n 19%|█▉        | 458/2376 [00:00<00:03, 504.50it/s]\n 21%|██▏       | 509/2376 [00:01<00:03, 503.92it/s]\n 24%|██▎       | 560/2376 [00:01<00:03, 499.55it/s]\n 26%|██▌       | 610/2376 [00:01<00:03, 497.58it/s]\n 28%|██▊       | 662/2376 [00:01<00:03, 501.66it/s]\n 30%|███       | 713/2376 [00:01<00:03, 498.48it/s]\n 32%|███▏      | 763/2376 [00:01<00:03, 496.29it/s]\n 34%|███▍      | 813/2376 [00:01<00:03, 497.24it/s]\n 36%|███▋      | 863/2376 [00:01<00:03, 489.91it/s]\n 38%|███▊      | 913/2376 [00:01<00:02, 490.66it/s]\n 41%|████      | 963/2376 [00:01<00:02, 480.89it/s]\n 43%|████▎     | 1012/2376 [00:02<00:02, 464.66it/s]\n 45%|████▍     | 1062/2376 [00:02<00:02, 473.99it/s]\n 47%|████▋     | 1112/2376 [00:02<00:02, 480.81it/s]\n 49%|████▉     | 1163/2376 [00:02<00:02, 487.90it/s]\n 51%|█████     | 1213/2376 [00:02<00:02, 488.65it/s]\n 53%|█████▎    | 1264/2376 [00:02<00:02, 492.04it/s]\n 55%|█████▌    | 1314/2376 [00:02<00:02, 491.65it/s]\n 57%|█████▋    | 1364/2376 [00:02<00:02, 493.50it/s]\n 60%|█████▉    | 1414/2376 [00:02<00:01, 490.67it/s]\n 62%|██████▏   | 1464/2376 [00:02<00:01, 487.53it/s]\n 64%|██████▎   | 1514/2376 [00:03<00:01, 489.47it/s]\n 66%|██████▌   | 1563/2376 [00:03<00:01, 486.54it/s]\n 68%|██████▊   | 1612/2376 [00:03<00:01, 487.03it/s]\n 70%|██████▉   | 1662/2376 [00:03<00:01, 488.58it/s]\n 72%|███████▏  | 1711/2376 [00:03<00:01, 486.13it/s]\n 74%|███████▍  | 1760/2376 [00:03<00:01, 486.07it/s]\n 76%|███████▌  | 1809/2376 [00:03<00:01, 486.57it/s]\n 78%|███████▊  | 1858/2376 [00:03<00:01, 487.30it/s]\n 80%|████████  | 1907/2376 [00:03<00:00, 486.60it/s]\n 82%|████████▏ | 1956/2376 [00:03<00:00, 487.11it/s]\n 84%|████████▍ | 2005/2376 [00:04<00:00, 485.56it/s]\n 86%|████████▋ | 2055/2376 [00:04<00:00, 487.25it/s]\n 89%|████████▊ | 2105/2376 [00:04<00:00, 489.42it/s]\n 91%|█████████ | 2155/2376 [00:04<00:00, 490.62it/s]\n 93%|█████████▎| 2205/2376 [00:04<00:00, 486.87it/s]\n 95%|█████████▍| 2256/2376 [00:04<00:00, 493.22it/s]\n 97%|█████████▋| 2307/2376 [00:04<00:00, 497.24it/s]\n 99%|█████████▉| 2359/2376 [00:04<00:00, 503.41it/s]\n100%|██████████| 2376/2376 [00:04<00:00, 492.10it/s]\n\nTokenizing inputs:   0%|          | 0/9501 [00:00<?, ?it/s]\nTokenizing inputs:   2%|▏         | 235/9501 [00:00<00:03, 2347.33it/s]\nTokenizing inputs:   5%|▍         | 470/9501 [00:00<00:03, 2300.45it/s]\nTokenizing inputs:   7%|▋         | 704/9501 [00:00<00:03, 2317.05it/s]\nTokenizing inputs:  10%|▉         | 936/9501 [00:00<00:03, 2266.05it/s]\nTokenizing inputs:  13%|█▎        | 1189/9501 [00:00<00:03, 2357.23it/s]\nTokenizing inputs:  15%|█▌        | 1438/9501 [00:00<00:03, 2398.96it/s]\nTokenizing inputs:  18%|█▊        | 1679/9501 [00:00<00:03, 2381.80it/s]\nTokenizing inputs:  20%|██        | 1926/9501 [00:00<00:03, 2409.00it/s]\nTokenizing inputs:  23%|██▎       | 2168/9501 [00:00<00:03, 2346.22it/s]\nTokenizing inputs:  25%|██▌       | 2404/9501 [00:01<00:03, 2331.95it/s]\nTokenizing inputs:  28%|██▊       | 2640/9501 [00:01<00:02, 2340.10it/s]\nTokenizing inputs:  30%|███       | 2891/9501 [00:01<00:02, 2390.81it/s]\nTokenizing inputs:  33%|███▎      | 3131/9501 [00:01<00:02, 2382.89it/s]\nTokenizing inputs:  36%|███▌      | 3378/9501 [00:01<00:02, 2406.69it/s]\nTokenizing inputs:  38%|███▊      | 3619/9501 [00:01<00:02, 2366.03it/s]\nTokenizing inputs:  41%|████      | 3856/9501 [00:01<00:02, 2267.41it/s]\nTokenizing inputs:  43%|████▎     | 4084/9501 [00:01<00:02, 2249.65it/s]\nTokenizing inputs:  46%|████▌     | 4323/9501 [00:01<00:02, 2288.15it/s]\nTokenizing inputs:  48%|████▊     | 4577/9501 [00:01<00:02, 2360.69it/s]\nTokenizing inputs:  51%|█████     | 4819/9501 [00:02<00:01, 2376.68it/s]\nTokenizing inputs:  53%|█████▎    | 5071/9501 [00:02<00:01, 2417.19it/s]\nTokenizing inputs:  56%|█████▌    | 5314/9501 [00:02<00:01, 2336.29it/s]\nTokenizing inputs:  59%|█████▊    | 5569/9501 [00:02<00:01, 2397.64it/s]\nTokenizing inputs:  61%|██████    | 5810/9501 [00:02<00:01, 2343.43it/s]\nTokenizing inputs:  64%|██████▎   | 6046/9501 [00:02<00:01, 2337.48it/s]\nTokenizing inputs:  66%|██████▋   | 6296/9501 [00:02<00:01, 2383.09it/s]\nTokenizing inputs:  69%|██████▉   | 6540/9501 [00:02<00:01, 2398.22it/s]\nTokenizing inputs:  71%|███████▏  | 6792/9501 [00:02<00:01, 2433.73it/s]\nTokenizing inputs:  74%|███████▍  | 7036/9501 [00:02<00:01, 2396.18it/s]\nTokenizing inputs:  77%|███████▋  | 7276/9501 [00:03<00:00, 2395.17it/s]\nTokenizing inputs:  79%|███████▉  | 7531/9501 [00:03<00:00, 2438.93it/s]\nTokenizing inputs:  82%|████████▏ | 7776/9501 [00:03<00:00, 2434.31it/s]\nTokenizing inputs:  84%|████████▍ | 8024/9501 [00:03<00:00, 2447.52it/s]\nTokenizing inputs:  87%|████████▋ | 8269/9501 [00:03<00:00, 2409.64it/s]\nTokenizing inputs:  90%|████████▉ | 8529/9501 [00:03<00:00, 2465.29it/s]\nTokenizing inputs:  92%|█████████▏| 8780/9501 [00:03<00:00, 2477.42it/s]\nTokenizing inputs:  95%|█████████▌| 9028/9501 [00:03<00:00, 2431.54it/s]\nTokenizing inputs:  98%|█████████▊| 9282/9501 [00:03<00:00, 2460.81it/s]\nTokenizing inputs: 100%|██████████| 9501/9501 [00:03<00:00, 2386.98it/s]\n\nRunning loglikelihood requests:   0%|          | 0/9501 [00:00<?, ?it/s]\nRunning loglikelihood requests:   0%|          | 1/9501 [00:12<32:27:51, 12.30s/it]\nRunning loglikelihood requests:   5%|▌         | 513/9501 [00:16<03:42, 40.41it/s] \nRunning loglikelihood requests:  11%|█         | 1025/9501 [00:19<02:00, 70.62it/s]\nRunning loglikelihood requests:  16%|█▌        | 1532/9501 [00:19<01:02, 127.87it/s]\nRunning loglikelihood requests:  19%|█▊        | 1772/9501 [00:22<01:09, 111.95it/s]\nRunning loglikelihood requests:  22%|██▏       | 2050/9501 [00:25<01:08, 108.33it/s]\nRunning loglikelihood requests:  27%|██▋       | 2559/9501 [00:25<00:37, 184.76it/s]\nRunning loglikelihood requests:  29%|██▉       | 2797/9501 [00:28<00:43, 152.61it/s]\nRunning loglikelihood requests:  32%|███▏      | 3077/9501 [00:30<00:45, 141.46it/s]\nRunning loglikelihood requests:  38%|███▊      | 3589/9501 [00:32<00:35, 168.31it/s]\nRunning loglikelihood requests:  43%|████▎     | 4102/9501 [00:35<00:28, 190.99it/s]\nRunning loglikelihood requests:  49%|████▊     | 4614/9501 [00:36<00:23, 211.57it/s]\nRunning loglikelihood requests:  54%|█████▍    | 5127/9501 [00:38<00:18, 232.37it/s]\nRunning loglikelihood requests:  59%|█████▉    | 5639/9501 [00:40<00:15, 252.14it/s]\nRunning loglikelihood requests:  65%|██████▍   | 6151/9501 [00:41<00:12, 271.31it/s]\nRunning loglikelihood requests:  70%|███████   | 6663/9501 [00:43<00:09, 291.66it/s]\nRunning loglikelihood requests:  76%|███████▌  | 7175/9501 [00:44<00:07, 312.87it/s]\nRunning loglikelihood requests:  81%|████████  | 7690/9501 [00:46<00:05, 336.46it/s]\nRunning loglikelihood requests:  86%|████████▌ | 8187/9501 [00:46<00:02, 465.12it/s]\nRunning loglikelihood requests:  88%|████████▊ | 8364/9501 [00:47<00:03, 369.20it/s]\nRunning loglikelihood requests:  92%|█████████▏| 8714/9501 [00:48<00:02, 357.28it/s]\nRunning loglikelihood requests:  97%|█████████▋| 9232/9501 [00:48<00:00, 462.15it/s]\nRunning loglikelihood requests: 100%|██████████| 9501/9501 [00:48<00:00, 194.12it/s]\nfatal: not a git repository (or any parent up to mount point /kaggle)\nStopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).\n{\n  \"arc_easy\": {\n    \"name\": \"arc_easy\",\n    \"alias\": \"arc_easy\",\n    \"sample_len\": 2376,\n    \"acc,none\": 0.5446127946127947,\n    \"acc_stderr,none\": 0.01021886178761889,\n    \"acc_norm,none\": 0.4730639730639731,\n    \"acc_norm_stderr,none\": 0.010244884740620078\n  }\n}\nVoid eval metadata: {\n  \"max_length\": 2048,\n  \"long_continuation_policy\": \"sliding_disjoint_segments\",\n  \"long_continuation_stride\": 256,\n  \"long_request_count\": 0,\n  \"long_continuation_tokens\": 0,\n  \"max_continuation_tokens\": 0\n}\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"  model           model_id                                  revision  \\\n0  Void  appvoid/void-byte  92c601560cc2179e3bfc33e067fdd3bf9254e163   \n\n   hellaswag       piqa   arc_easy  \n0  38.677554  67.464635  47.306397  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>model</th>\n      <th>model_id</th>\n      <th>revision</th>\n      <th>hellaswag</th>\n      <th>piqa</th>\n      <th>arc_easy</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Void</td>\n      <td>appvoid/void-byte</td>\n      <td>92c601560cc2179e3bfc33e067fdd3bf9254e163</td>\n      <td>38.677554</td>\n      <td>67.464635</td>\n      <td>47.306397</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"\n================================================================================\nVOID — arc_challenge\n================================================================================\n\n[arc_challenge] trying fixed batch=512\n\nCOMMAND: /usr/bin/python3 /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_run_void_lm_eval.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --revision 92c601560cc2179e3bfc33e067fdd3bf9254e163 --task arc_challenge --device cuda:0 --dtype float16 --batch-size 512 --max-length 2048 --long-cont-stride 256 --out /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/open_slm/arc_challenge/results.json\ntask = arc_challenge batch = 512 max_length = 2048 long_cont_stride = 256\n\nGenerating train split:   0%|          | 0/1119 [00:00<?, ? examples/s]\nGenerating train split: 100%|██████████| 1119/1119 [00:00<00:00, 162892.66 examples/s]\n\nGenerating test split:   0%|          | 0/1172 [00:00<?, ? examples/s]\nGenerating test split: 100%|██████████| 1172/1172 [00:00<00:00, 189613.28 examples/s]\n\nGenerating validation split:   0%|          | 0/299 [00:00<?, ? examples/s]\nGenerating validation split: 100%|██████████| 299/299 [00:00<00:00, 88584.93 examples/s]\nWARNING:lm_eval.evaluator:Overwriting default num_fewshot of arc_challenge from None to 0\n\n  0%|          | 0/1172 [00:00<?, ?it/s]\n  4%|▍         | 48/1172 [00:00<00:02, 473.47it/s]\n  8%|▊         | 96/1172 [00:00<00:02, 461.86it/s]\n 12%|█▏        | 145/1172 [00:00<00:02, 471.85it/s]\n 16%|█▋        | 193/1172 [00:00<00:02, 472.77it/s]\n 21%|██        | 241/1172 [00:00<00:01, 468.81it/s]\n 25%|██▍       | 288/1172 [00:00<00:01, 468.13it/s]\n 29%|██▊       | 335/1172 [00:00<00:01, 455.96it/s]\n 33%|███▎      | 385/1172 [00:00<00:01, 468.82it/s]\n 37%|███▋      | 438/1172 [00:00<00:01, 485.31it/s]\n 42%|████▏     | 490/1172 [00:01<00:01, 495.57it/s]\n 46%|████▌     | 541/1172 [00:01<00:01, 499.62it/s]\n 51%|█████     | 592/1172 [00:01<00:01, 502.16it/s]\n 55%|█████▍    | 644/1172 [00:01<00:01, 504.79it/s]\n 59%|█████▉    | 696/1172 [00:01<00:00, 508.15it/s]\n 64%|██████▎   | 747/1172 [00:01<00:00, 508.49it/s]\n 68%|██████▊   | 799/1172 [00:01<00:00, 510.58it/s]\n 73%|███████▎  | 851/1172 [00:01<00:00, 512.10it/s]\n 77%|███████▋  | 903/1172 [00:01<00:00, 510.56it/s]\n 81%|████████▏ | 955/1172 [00:01<00:00, 509.41it/s]\n 86%|████████▌ | 1006/1172 [00:02<00:00, 509.29it/s]\n 90%|█████████ | 1058/1172 [00:02<00:00, 512.33it/s]\n 95%|█████████▍| 1110/1172 [00:02<00:00, 512.37it/s]\n 99%|█████████▉| 1162/1172 [00:02<00:00, 513.05it/s]\n100%|██████████| 1172/1172 [00:02<00:00, 497.07it/s]\n\nTokenizing inputs:   0%|          | 0/4687 [00:00<?, ?it/s]\nTokenizing inputs:   4%|▍         | 201/4687 [00:00<00:02, 2000.32it/s]\nTokenizing inputs:   9%|▉         | 415/4687 [00:00<00:02, 2082.06it/s]\nTokenizing inputs:  14%|█▍        | 656/4687 [00:00<00:01, 2231.44it/s]\nTokenizing inputs:  19%|█▉        | 880/4687 [00:00<00:01, 2205.91it/s]\nTokenizing inputs:  23%|██▎       | 1101/4687 [00:00<00:01, 2123.94it/s]\nTokenizing inputs:  28%|██▊       | 1314/4687 [00:00<00:01, 2103.71it/s]\nTokenizing inputs:  33%|███▎      | 1531/4687 [00:00<00:01, 2123.71it/s]\nTokenizing inputs:  37%|███▋      | 1744/4687 [00:00<00:01, 2000.95it/s]\nTokenizing inputs:  42%|████▏     | 1946/4687 [00:00<00:01, 1998.54it/s]\nTokenizing inputs:  46%|████▌     | 2165/4687 [00:01<00:01, 2033.65it/s]\nTokenizing inputs:  51%|█████     | 2370/4687 [00:01<00:01, 2028.37it/s]\nTokenizing inputs:  55%|█████▍    | 2575/4687 [00:01<00:01, 2032.10it/s]\nTokenizing inputs:  59%|█████▉    | 2779/4687 [00:01<00:00, 1948.72it/s]\nTokenizing inputs:  64%|██████▍   | 2993/4687 [00:01<00:00, 2001.88it/s]\nTokenizing inputs:  68%|██████▊   | 3208/4687 [00:01<00:00, 2043.91it/s]\nTokenizing inputs:  73%|███████▎  | 3428/4687 [00:01<00:00, 2089.39it/s]\nTokenizing inputs:  78%|███████▊  | 3638/4687 [00:01<00:00, 2086.70it/s]\nTokenizing inputs:  82%|████████▏ | 3858/4687 [00:01<00:00, 2119.74it/s]\nTokenizing inputs:  87%|████████▋ | 4076/4687 [00:01<00:00, 2137.37it/s]\nTokenizing inputs:  92%|█████████▏| 4290/4687 [00:02<00:00, 2114.12it/s]\nTokenizing inputs:  96%|█████████▌| 4505/4687 [00:02<00:00, 2123.12it/s]\nTokenizing inputs: 100%|██████████| 4687/4687 [00:02<00:00, 2085.90it/s]\n\nRunning loglikelihood requests:   0%|          | 0/4687 [00:00<?, ?it/s]\nRunning loglikelihood requests:   0%|          | 1/4687 [00:13<17:37:44, 13.54s/it]\nRunning loglikelihood requests:  11%|█         | 506/4687 [00:13<01:19, 52.78it/s] \nRunning loglikelihood requests:  22%|██▏       | 1010/4687 [00:17<00:45, 81.54it/s]\nRunning loglikelihood requests:  26%|██▌       | 1226/4687 [00:20<00:44, 77.78it/s]\nRunning loglikelihood requests:  33%|███▎      | 1547/4687 [00:23<00:35, 88.14it/s]\nRunning loglikelihood requests:  44%|████▍     | 2059/4687 [00:25<00:22, 118.83it/s]\nRunning loglikelihood requests:  54%|█████▍    | 2544/4687 [00:25<00:11, 188.43it/s]\nRunning loglikelihood requests:  59%|█████▊    | 2747/4687 [00:28<00:12, 161.13it/s]\nRunning loglikelihood requests:  66%|██████▌   | 3091/4687 [00:29<00:09, 169.22it/s]\nRunning loglikelihood requests:  76%|███████▋  | 3580/4687 [00:29<00:04, 269.51it/s]\nRunning loglikelihood requests:  81%|████████  | 3803/4687 [00:31<00:03, 227.95it/s]\nRunning loglikelihood requests:  88%|████████▊ | 4121/4687 [00:32<00:02, 229.72it/s]\nRunning loglikelihood requests:  99%|█████████▊| 4626/4687 [00:32<00:00, 369.89it/s]\nRunning loglikelihood requests: 100%|██████████| 4687/4687 [00:33<00:00, 141.83it/s]\nfatal: not a git repository (or any parent up to mount point /kaggle)\nStopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).\n{\n  \"arc_challenge\": {\n    \"name\": \"arc_challenge\",\n    \"alias\": \"arc_challenge\",\n    \"sample_len\": 1172,\n    \"acc,none\": 0.24829351535836178,\n    \"acc_stderr,none\": 0.012624912868089679,\n    \"acc_norm,none\": 0.2815699658703072,\n    \"acc_norm_stderr,none\": 0.0131433767350091\n  }\n}\nVoid eval metadata: {\n  \"max_length\": 2048,\n  \"long_continuation_policy\": \"sliding_disjoint_segments\",\n  \"long_continuation_stride\": 256,\n  \"long_request_count\": 0,\n  \"long_continuation_tokens\": 0,\n  \"max_continuation_tokens\": 0\n}\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"  model           model_id                                  revision  \\\n0  Void  appvoid/void-byte  92c601560cc2179e3bfc33e067fdd3bf9254e163   \n\n   hellaswag       piqa   arc_easy  arc_challenge  \n0  38.677554  67.464635  47.306397      28.156997  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>model</th>\n      <th>model_id</th>\n      <th>revision</th>\n      <th>hellaswag</th>\n      <th>piqa</th>\n      <th>arc_easy</th>\n      <th>arc_challenge</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Void</td>\n      <td>appvoid/void-byte</td>\n      <td>92c601560cc2179e3bfc33e067fdd3bf9254e163</td>\n      <td>38.677554</td>\n      <td>67.464635</td>\n      <td>47.306397</td>\n      <td>28.156997</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"\nCompleted all requested Open SLM tasks.\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"  model           model_id                                  revision  \\\n0  Void  appvoid/void-byte  92c601560cc2179e3bfc33e067fdd3bf9254e163   \n\n   hellaswag       piqa   arc_easy  arc_challenge  \n0  38.677554  67.464635  47.306397      28.156997  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>model</th>\n      <th>model_id</th>\n      <th>revision</th>\n      <th>hellaswag</th>\n      <th>piqa</th>\n      <th>arc_easy</th>\n      <th>arc_challenge</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Void</td>\n      <td>appvoid/void-byte</td>\n      <td>92c601560cc2179e3bfc33e067fdd3bf9254e163</td>\n      <td>38.677554</td>\n      <td>67.464635</td>\n      <td>47.306397</td>\n      <td>28.156997</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"Combined summary: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/open_slm/summary.json\n\nCOMMAND: /usr/bin/python3 /root/.cache/huggingface/hub/datasets--BananaMind--BananaMind-Base-Bench-1.1/snapshots/d4aade51312889e8580963e1ce960c6eaef1a450/benchmark.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --model-revision 92c601560cc2179e3bfc33e067fdd3bf9254e163 --dataset-revision d4aade51312889e8580963e1ce960c6eaef1a450 --device cuda --dtype float16 --batch-size 32 --threads 2 --out-dir /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1 --max-context 2048\nChecking --hf-token/HF_TOKEN access to BananaMind/BananaMind-Base-Bench-1.1...\nDataset access confirmed.\nLoading /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model on cuda as float16...\nScoring raw continuations with context length 2048; add_bos=False.\n[001/350] PASS language_completion-001 (pred=2, label=2)\n[002/350] PASS language_completion-002 (pred=1, label=1)\n[003/350] PASS language_completion-003 (pred=2, label=2)\n[004/350] PASS language_completion-004 (pred=3, label=3)\n[005/350] PASS language_completion-005 (pred=0, label=0)\n[006/350] PASS language_completion-006 (pred=1, label=1)\n[007/350] PASS language_completion-007 (pred=2, label=2)\n[008/350] PASS language_completion-008 (pred=3, label=3)\n[009/350] PASS language_completion-009 (pred=0, label=0)\n[010/350] PASS language_completion-010 (pred=1, label=1)\n[011/350] PASS language_completion-011 (pred=2, label=2)\n[012/350] PASS language_completion-012 (pred=3, label=3)\n[013/350] PASS language_completion-013 (pred=0, label=0)\n[014/350] PASS language_completion-014 (pred=1, label=1)\n[015/350] PASS language_completion-015 (pred=2, label=2)\n[016/350] FAIL language_completion-016 (pred=0, label=3)\n[017/350] PASS language_completion-017 (pred=0, label=0)\n[018/350] FAIL language_completion-018 (pred=0, label=1)\n[019/350] PASS language_completion-019 (pred=2, label=2)\n[020/350] PASS language_completion-020 (pred=3, label=3)\n[021/350] PASS language_completion-021 (pred=0, label=0)\n[022/350] PASS language_completion-022 (pred=1, label=1)\n[023/350] PASS language_completion-023 (pred=2, label=2)\n[024/350] PASS language_completion-024 (pred=3, label=3)\n[025/350] FAIL language_completion-025 (pred=3, label=0)\n[026/350] PASS language_completion-026 (pred=1, label=1)\n[027/350] PASS language_completion-027 (pred=2, label=2)\n[028/350] PASS language_completion-028 (pred=3, label=3)\n[029/350] PASS language_completion-029 (pred=0, label=0)\n[030/350] PASS language_completion-030 (pred=1, label=1)\n[031/350] PASS language_completion-031 (pred=2, label=2)\n[032/350] FAIL language_completion-032 (pred=0, label=3)\n[033/350] PASS language_completion-033 (pred=0, label=0)\n[034/350] PASS language_completion-034 (pred=1, label=1)\n[035/350] PASS language_completion-035 (pred=2, label=2)\n[036/350] PASS language_completion-036 (pred=3, label=3)\n[037/350] PASS language_completion-037 (pred=0, label=0)\n[038/350] PASS language_completion-038 (pred=1, label=1)\n[039/350] PASS language_completion-039 (pred=2, label=2)\n[040/350] PASS language_completion-040 (pred=3, label=3)\n[041/350] PASS language_completion-041 (pred=0, label=0)\n[042/350] FAIL language_completion-042 (pred=2, label=1)\n[043/350] PASS language_completion-043 (pred=2, label=2)\n[044/350] PASS language_completion-044 (pred=3, label=3)\n[045/350] PASS language_completion-045 (pred=0, label=0)\n[046/350] PASS language_completion-046 (pred=1, label=1)\n[047/350] PASS language_completion-047 (pred=2, label=2)\n[048/350] PASS language_completion-048 (pred=3, label=3)\n[049/350] PASS language_completion-049 (pred=0, label=0)\n[050/350] PASS language_completion-050 (pred=1, label=1)\n[051/350] PASS commonsense-001 (pred=2, label=2)\n[052/350] PASS commonsense-002 (pred=1, label=1)\n[053/350] PASS commonsense-003 (pred=2, label=2)\n[054/350] PASS commonsense-004 (pred=3, label=3)\n[055/350] PASS commonsense-005 (pred=0, label=0)\n[056/350] PASS commonsense-006 (pred=1, label=1)\n[057/350] PASS commonsense-007 (pred=2, label=2)\n[058/350] FAIL commonsense-008 (pred=0, label=3)\n[059/350] PASS commonsense-009 (pred=0, label=0)\n[060/350] PASS commonsense-010 (pred=1, label=1)\n[061/350] PASS commonsense-011 (pred=2, label=2)\n[062/350] PASS commonsense-012 (pred=3, label=3)\n[063/350] PASS commonsense-013 (pred=0, label=0)\n[064/350] PASS commonsense-014 (pred=1, label=1)\n[065/350] PASS commonsense-015 (pred=2, label=2)\n[066/350] PASS commonsense-016 (pred=3, label=3)\n[067/350] PASS commonsense-017 (pred=0, label=0)\n[068/350] PASS commonsense-018 (pred=1, label=1)\n[069/350] PASS commonsense-019 (pred=2, label=2)\n[070/350] PASS commonsense-020 (pred=3, label=3)\n[071/350] PASS commonsense-021 (pred=0, label=0)\n[072/350] PASS commonsense-022 (pred=1, label=1)\n[073/350] FAIL commonsense-023 (pred=3, label=2)\n[074/350] FAIL commonsense-024 (pred=0, label=3)\n[075/350] PASS commonsense-025 (pred=0, label=0)\n[076/350] FAIL commonsense-026 (pred=2, label=1)\n[077/350] PASS commonsense-027 (pred=2, label=2)\n[078/350] PASS commonsense-028 (pred=3, label=3)\n[079/350] FAIL commonsense-029 (pred=3, label=0)\n[080/350] PASS commonsense-030 (pred=1, label=1)\n[081/350] PASS commonsense-031 (pred=2, label=2)\n[082/350] FAIL commonsense-032 (pred=0, label=3)\n[083/350] FAIL commonsense-033 (pred=1, label=0)\n[084/350] PASS commonsense-034 (pred=1, label=1)\n[085/350] FAIL commonsense-035 (pred=3, label=2)\n[086/350] PASS commonsense-036 (pred=3, label=3)\n[087/350] PASS commonsense-037 (pred=0, label=0)\n[088/350] PASS commonsense-038 (pred=1, label=1)\n[089/350] FAIL commonsense-039 (pred=0, label=2)\n[090/350] PASS commonsense-040 (pred=3, label=3)\n[091/350] PASS commonsense-041 (pred=0, label=0)\n[092/350] FAIL commonsense-042 (pred=0, label=1)\n[093/350] PASS commonsense-043 (pred=2, label=2)\n[094/350] PASS commonsense-044 (pred=3, label=3)\n[095/350] PASS commonsense-045 (pred=0, label=0)\n[096/350] FAIL commonsense-046 (pred=0, label=1)\n[097/350] PASS commonsense-047 (pred=2, label=2)\n[098/350] PASS commonsense-048 (pred=3, label=3)\n[099/350] FAIL commonsense-049 (pred=2, label=0)\n[100/350] PASS commonsense-050 (pred=1, label=1)\n[101/350] PASS world_knowledge-001 (pred=2, label=2)\n[102/350] PASS world_knowledge-002 (pred=1, label=1)\n[103/350] FAIL world_knowledge-003 (pred=1, label=2)\n[104/350] PASS world_knowledge-004 (pred=3, label=3)\n[105/350] PASS world_knowledge-005 (pred=0, label=0)\n[106/350] PASS world_knowledge-006 (pred=1, label=1)\n[107/350] PASS world_knowledge-007 (pred=2, label=2)\n[108/350] PASS world_knowledge-008 (pred=3, label=3)\n[109/350] PASS world_knowledge-009 (pred=0, label=0)\n[110/350] PASS world_knowledge-010 (pred=1, label=1)\n[111/350] FAIL world_knowledge-011 (pred=0, label=2)\n[112/350] FAIL world_knowledge-012 (pred=2, label=3)\n[113/350] PASS world_knowledge-013 (pred=0, label=0)\n[114/350] PASS world_knowledge-014 (pred=1, label=1)\n[115/350] PASS world_knowledge-015 (pred=2, label=2)\n[116/350] PASS world_knowledge-016 (pred=3, label=3)\n[117/350] PASS world_knowledge-017 (pred=0, label=0)\n[118/350] PASS world_knowledge-018 (pred=1, label=1)\n[119/350] FAIL world_knowledge-019 (pred=1, label=2)\n[120/350] PASS world_knowledge-020 (pred=3, label=3)\n[121/350] PASS world_knowledge-021 (pred=0, label=0)\n[122/350] PASS world_knowledge-022 (pred=1, label=1)\n[123/350] PASS world_knowledge-023 (pred=2, label=2)\n[124/350] PASS world_knowledge-024 (pred=3, label=3)\n[125/350] PASS world_knowledge-025 (pred=0, label=0)\n[126/350] PASS world_knowledge-026 (pred=1, label=1)\n[127/350] FAIL world_knowledge-027 (pred=1, label=2)\n[128/350] PASS world_knowledge-028 (pred=3, label=3)\n[129/350] PASS world_knowledge-029 (pred=0, label=0)\n[130/350] PASS world_knowledge-030 (pred=1, label=1)\n[131/350] PASS world_knowledge-031 (pred=2, label=2)\n[132/350] FAIL world_knowledge-032 (pred=1, label=3)\n[133/350] PASS world_knowledge-033 (pred=0, label=0)\n[134/350] PASS world_knowledge-034 (pred=1, label=1)\n[135/350] PASS world_knowledge-035 (pred=2, label=2)\n[136/350] PASS world_knowledge-036 (pred=3, label=3)\n[137/350] PASS world_knowledge-037 (pred=0, label=0)\n[138/350] PASS world_knowledge-038 (pred=1, label=1)\n[139/350] FAIL world_knowledge-039 (pred=1, label=2)\n[140/350] PASS world_knowledge-040 (pred=3, label=3)\n[141/350] FAIL world_knowledge-041 (pred=2, label=0)\n[142/350] FAIL world_knowledge-042 (pred=0, label=1)\n[143/350] PASS world_knowledge-043 (pred=2, label=2)\n[144/350] PASS world_knowledge-044 (pred=3, label=3)\n[145/350] FAIL world_knowledge-045 (pred=1, label=0)\n[146/350] FAIL world_knowledge-046 (pred=0, label=1)\n[147/350] PASS world_knowledge-047 (pred=2, label=2)\n[148/350] FAIL world_knowledge-048 (pred=0, label=3)\n[149/350] PASS world_knowledge-049 (pred=0, label=0)\n[150/350] FAIL world_knowledge-050 (pred=0, label=1)\n[151/350] PASS context_tracking-001 (pred=0, label=0)\n[152/350] FAIL context_tracking-002 (pred=0, label=3)\n[153/350] PASS context_tracking-003 (pred=2, label=2)\n[154/350] PASS context_tracking-004 (pred=3, label=3)\n[155/350] FAIL context_tracking-005 (pred=1, label=0)\n[156/350] FAIL context_tracking-006 (pred=3, label=1)\n[157/350] PASS context_tracking-007 (pred=2, label=2)\n[158/350] FAIL context_tracking-008 (pred=1, label=3)\n[159/350] PASS context_tracking-009 (pred=0, label=0)\n[160/350] FAIL context_tracking-010 (pred=2, label=1)\n[161/350] FAIL context_tracking-011 (pred=1, label=2)\n[162/350] FAIL context_tracking-012 (pred=2, label=3)\n[163/350] PASS context_tracking-013 (pred=0, label=0)\n[164/350] PASS context_tracking-014 (pred=1, label=1)\n[165/350] PASS context_tracking-015 (pred=2, label=2)\n[166/350] PASS context_tracking-016 (pred=3, label=3)\n[167/350] FAIL context_tracking-017 (pred=1, label=0)\n[168/350] FAIL context_tracking-018 (pred=0, label=1)\n[169/350] FAIL context_tracking-019 (pred=1, label=2)\n[170/350] PASS context_tracking-020 (pred=3, label=3)\n[171/350] FAIL context_tracking-021 (pred=1, label=0)\n[172/350] FAIL context_tracking-022 (pred=0, label=1)\n[173/350] PASS context_tracking-023 (pred=2, label=2)\n[174/350] FAIL context_tracking-024 (pred=0, label=3)\n[175/350] FAIL context_tracking-025 (pred=1, label=0)\n[176/350] FAIL context_tracking-026 (pred=3, label=1)\n[177/350] FAIL context_tracking-027 (pred=0, label=2)\n[178/350] FAIL context_tracking-028 (pred=0, label=3)\n[179/350] FAIL context_tracking-029 (pred=3, label=0)\n[180/350] PASS context_tracking-030 (pred=1, label=1)\n[181/350] FAIL context_tracking-031 (pred=0, label=2)\n[182/350] FAIL context_tracking-032 (pred=1, label=3)\n[183/350] FAIL context_tracking-033 (pred=1, label=0)\n[184/350] PASS context_tracking-034 (pred=1, label=1)\n[185/350] FAIL context_tracking-035 (pred=0, label=2)\n[186/350] PASS context_tracking-036 (pred=3, label=3)\n[187/350] FAIL context_tracking-037 (pred=2, label=0)\n[188/350] PASS context_tracking-038 (pred=1, label=1)\n[189/350] PASS context_tracking-039 (pred=2, label=2)\n[190/350] PASS context_tracking-040 (pred=3, label=3)\n[191/350] FAIL context_tracking-041 (pred=2, label=0)\n[192/350] FAIL context_tracking-042 (pred=2, label=1)\n[193/350] FAIL context_tracking-043 (pred=1, label=2)\n[194/350] FAIL context_tracking-044 (pred=1, label=3)\n[195/350] FAIL context_tracking-045 (pred=3, label=0)\n[196/350] FAIL context_tracking-046 (pred=3, label=1)\n[197/350] FAIL context_tracking-047 (pred=0, label=2)\n[198/350] PASS context_tracking-048 (pred=3, label=3)\n[199/350] FAIL context_tracking-049 (pred=1, label=0)\n[200/350] FAIL context_tracking-050 (pred=3, label=1)\n[201/350] FAIL quantitative-001 (pred=2, label=0)\n[202/350] FAIL quantitative-002 (pred=2, label=3)\n[203/350] PASS quantitative-003 (pred=2, label=2)\n[204/350] FAIL quantitative-004 (pred=1, label=3)\n[205/350] PASS quantitative-005 (pred=0, label=0)\n[206/350] FAIL quantitative-006 (pred=0, label=1)\n[207/350] FAIL quantitative-007 (pred=3, label=2)\n[208/350] FAIL quantitative-008 (pred=0, label=3)\n[209/350] FAIL quantitative-009 (pred=2, label=0)\n[210/350] PASS quantitative-010 (pred=1, label=1)\n[211/350] PASS quantitative-011 (pred=2, label=2)\n[212/350] PASS quantitative-012 (pred=3, label=3)\n[213/350] FAIL quantitative-013 (pred=3, label=0)\n[214/350] FAIL quantitative-014 (pred=0, label=1)\n[215/350] FAIL quantitative-015 (pred=3, label=2)\n[216/350] FAIL quantitative-016 (pred=2, label=3)\n[217/350] PASS quantitative-017 (pred=0, label=0)\n[218/350] FAIL quantitative-018 (pred=0, label=1)\n[219/350] FAIL quantitative-019 (pred=1, label=2)\n[220/350] PASS quantitative-020 (pred=3, label=3)\n[221/350] FAIL quantitative-021 (pred=2, label=0)\n[222/350] FAIL quantitative-022 (pred=3, label=1)\n[223/350] FAIL quantitative-023 (pred=3, label=2)\n[224/350] FAIL quantitative-024 (pred=0, label=3)\n[225/350] PASS quantitative-025 (pred=0, label=0)\n[226/350] FAIL quantitative-026 (pred=0, label=1)\n[227/350] PASS quantitative-027 (pred=2, label=2)\n[228/350] FAIL quantitative-028 (pred=2, label=3)\n[229/350] FAIL quantitative-029 (pred=3, label=0)\n[230/350] PASS quantitative-030 (pred=1, label=1)\n[231/350] FAIL quantitative-031 (pred=1, label=2)\n[232/350] FAIL quantitative-032 (pred=2, label=3)\n[233/350] FAIL quantitative-033 (pred=3, label=0)\n[234/350] FAIL quantitative-034 (pred=0, label=1)\n[235/350] FAIL quantitative-035 (pred=0, label=2)\n[236/350] FAIL quantitative-036 (pred=1, label=3)\n[237/350] FAIL quantitative-037 (pred=1, label=0)\n[238/350] FAIL quantitative-038 (pred=0, label=1)\n[239/350] FAIL quantitative-039 (pred=3, label=2)\n[240/350] FAIL quantitative-040 (pred=2, label=3)\n[241/350] FAIL quantitative-041 (pred=1, label=0)\n[242/350] FAIL quantitative-042 (pred=0, label=1)\n[243/350] PASS quantitative-043 (pred=2, label=2)\n[244/350] PASS quantitative-044 (pred=3, label=3)\n[245/350] FAIL quantitative-045 (pred=1, label=0)\n[246/350] FAIL quantitative-046 (pred=2, label=1)\n[247/350] FAIL quantitative-047 (pred=3, label=2)\n[248/350] FAIL quantitative-048 (pred=0, label=3)\n[249/350] FAIL quantitative-049 (pred=3, label=0)\n[250/350] PASS quantitative-050 (pred=1, label=1)\n[251/350] PASS logical_reasoning-001 (pred=0, label=0)\n[252/350] FAIL logical_reasoning-002 (pred=0, label=3)\n[253/350] PASS logical_reasoning-003 (pred=2, label=2)\n[254/350] FAIL logical_reasoning-004 (pred=0, label=3)\n[255/350] PASS logical_reasoning-005 (pred=0, label=0)\n[256/350] PASS logical_reasoning-006 (pred=1, label=1)\n[257/350] PASS logical_reasoning-007 (pred=2, label=2)\n[258/350] PASS logical_reasoning-008 (pred=3, label=3)\n[259/350] PASS logical_reasoning-009 (pred=0, label=0)\n[260/350] PASS logical_reasoning-010 (pred=1, label=1)\n[261/350] FAIL logical_reasoning-011 (pred=3, label=2)\n[262/350] FAIL logical_reasoning-012 (pred=2, label=3)\n[263/350] FAIL logical_reasoning-013 (pred=2, label=0)\n[264/350] PASS logical_reasoning-014 (pred=1, label=1)\n[265/350] PASS logical_reasoning-015 (pred=2, label=2)\n[266/350] FAIL logical_reasoning-016 (pred=2, label=3)\n[267/350] FAIL logical_reasoning-017 (pred=1, label=0)\n[268/350] PASS logical_reasoning-018 (pred=1, label=1)\n[269/350] PASS logical_reasoning-019 (pred=2, label=2)\n[270/350] FAIL logical_reasoning-020 (pred=0, label=3)\n[271/350] PASS logical_reasoning-021 (pred=0, label=0)\n[272/350] FAIL logical_reasoning-022 (pred=0, label=1)\n[273/350] FAIL logical_reasoning-023 (pred=0, label=2)\n[274/350] FAIL logical_reasoning-024 (pred=2, label=3)\n[275/350] PASS logical_reasoning-025 (pred=0, label=0)\n[276/350] FAIL logical_reasoning-026 (pred=2, label=1)\n[277/350] FAIL logical_reasoning-027 (pred=0, label=2)\n[278/350] FAIL logical_reasoning-028 (pred=0, label=3)\n[279/350] PASS logical_reasoning-029 (pred=0, label=0)\n[280/350] PASS logical_reasoning-030 (pred=1, label=1)\n[281/350] FAIL logical_reasoning-031 (pred=1, label=2)\n[282/350] FAIL logical_reasoning-032 (pred=2, label=3)\n[283/350] FAIL logical_reasoning-033 (pred=1, label=0)\n[284/350] FAIL logical_reasoning-034 (pred=2, label=1)\n[285/350] FAIL logical_reasoning-035 (pred=0, label=2)\n[286/350] FAIL logical_reasoning-036 (pred=2, label=3)\n[287/350] PASS logical_reasoning-037 (pred=0, label=0)\n[288/350] FAIL logical_reasoning-038 (pred=2, label=1)\n[289/350] FAIL logical_reasoning-039 (pred=0, label=2)\n[290/350] FAIL logical_reasoning-040 (pred=2, label=3)\n[291/350] FAIL logical_reasoning-041 (pred=1, label=0)\n[292/350] PASS logical_reasoning-042 (pred=1, label=1)\n[293/350] PASS logical_reasoning-043 (pred=2, label=2)\n[294/350] FAIL logical_reasoning-044 (pred=0, label=3)\n[295/350] PASS logical_reasoning-045 (pred=0, label=0)\n[296/350] FAIL logical_reasoning-046 (pred=2, label=1)\n[297/350] PASS logical_reasoning-047 (pred=2, label=2)\n[298/350] FAIL logical_reasoning-048 (pred=2, label=3)\n[299/350] FAIL logical_reasoning-049 (pred=2, label=0)\n[300/350] PASS logical_reasoning-050 (pred=1, label=1)\n[301/350] PASS code_completion-001 (pred=0, label=0)\n[302/350] FAIL code_completion-002 (pred=0, label=1)\n[303/350] PASS code_completion-003 (pred=2, label=2)\n[304/350] FAIL code_completion-004 (pred=2, label=3)\n[305/350] PASS code_completion-005 (pred=0, label=0)\n[306/350] FAIL code_completion-006 (pred=0, label=1)\n[307/350] FAIL code_completion-007 (pred=0, label=2)\n[308/350] PASS code_completion-008 (pred=3, label=3)\n[309/350] FAIL code_completion-009 (pred=3, label=0)\n[310/350] FAIL code_completion-010 (pred=2, label=1)\n[311/350] FAIL code_completion-011 (pred=0, label=2)\n[312/350] FAIL code_completion-012 (pred=2, label=3)\n[313/350] PASS code_completion-013 (pred=0, label=0)\n[314/350] PASS code_completion-014 (pred=1, label=1)\n[315/350] FAIL code_completion-015 (pred=3, label=2)\n[316/350] FAIL code_completion-016 (pred=1, label=3)\n[317/350] FAIL code_completion-017 (pred=3, label=0)\n[318/350] FAIL code_completion-018 (pred=0, label=1)\n[319/350] FAIL code_completion-019 (pred=1, label=2)\n[320/350] PASS code_completion-020 (pred=3, label=3)\n[321/350] PASS code_completion-021 (pred=0, label=0)\n[322/350] FAIL code_completion-022 (pred=0, label=1)\n[323/350] PASS code_completion-023 (pred=2, label=2)\n[324/350] FAIL code_completion-024 (pred=1, label=3)\n[325/350] FAIL code_completion-025 (pred=1, label=0)\n[326/350] FAIL code_completion-026 (pred=2, label=1)\n[327/350] FAIL code_completion-027 (pred=0, label=2)\n[328/350] FAIL code_completion-028 (pred=2, label=3)\n[329/350] FAIL code_completion-029 (pred=2, label=0)\n[330/350] FAIL code_completion-030 (pred=3, label=1)\n[331/350] FAIL code_completion-031 (pred=3, label=2)\n[332/350] FAIL code_completion-032 (pred=0, label=3)\n[333/350] FAIL code_completion-033 (pred=1, label=0)\n[334/350] FAIL code_completion-034 (pred=0, label=1)\n[335/350] FAIL code_completion-035 (pred=1, label=2)\n[336/350] FAIL code_completion-036 (pred=0, label=3)\n[337/350] FAIL code_completion-037 (pred=1, label=0)\n[338/350] FAIL code_completion-038 (pred=0, label=1)\n[339/350] FAIL code_completion-039 (pred=1, label=2)\n[340/350] FAIL code_completion-040 (pred=2, label=3)\n[341/350] PASS code_completion-041 (pred=0, label=0)\n[342/350] PASS code_completion-042 (pred=1, label=1)\n[343/350] PASS code_completion-043 (pred=2, label=2)\n[344/350] PASS code_completion-044 (pred=3, label=3)\n[345/350] PASS code_completion-045 (pred=0, label=0)\n[346/350] FAIL code_completion-046 (pred=0, label=1)\n[347/350] FAIL code_completion-047 (pred=0, label=2)\n[348/350] FAIL code_completion-048 (pred=2, label=3)\n[349/350] FAIL code_completion-049 (pred=2, label=0)\n[350/350] FAIL code_completion-050 (pred=3, label=1)\n\nBananaMind Base Bench 1.1\nOverall Elo: 1007\nAccuracy: 187/350 (53.43%)\nWeighted accuracy: 48.92%\nlanguage_completion: Elo 1273 | 45/50 (90.00%) | weighted 90.13%\ncommonsense: Elo 1132 | 38/50 (76.00%) | weighted 73.19%\nworld_knowledge: Elo 1112 | 37/50 (74.00%) | weighted 70.85%\ncontext_tracking: Elo 874 | 18/50 (36.00%) | weighted 33.44%\nquantitative: Elo 840 | 13/50 (26.00%) | weighted 23.66%\nlogical_reasoning: Elo 1031 | 22/50 (44.00%) | weighted 40.75%\ncode_completion: Elo 921 | 14/50 (28.00%) | weighted 27.71%\nReport: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1/report.json\nPredictions: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1/predictions.md\nResult JSON: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/bananamind_base_1_1/report.json\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"   elo.scale  elo.prior_rating  elo.prior_weight  summary.accuracy  \\\n0      400.0            1000.0               4.0          0.534286   \n\n   summary.weighted_points  summary.possible_weighted_points  \\\n0                    326.9                          668.1875   \n\n   summary.weighted_accuracy  summary.overall_elo  \\\n0                   0.489234                 1007   \n\n   summary.overall_elo_unrounded  \\\n0                    1007.356924   \n\n   summary.categories.language_completion.accuracy  ...  \\\n0                                              0.9  ...   \n\n   results[340].item_elo  results[341].item_elo  results[342].item_elo  \\\n0                   1200                   1200                   1200   \n\n   results[343].item_elo  results[344].item_elo  results[345].item_elo  \\\n0                   1200                   1200                   1200   \n\n   results[346].item_elo  results[347].item_elo  results[348].item_elo  \\\n0                   1200                   1200                   1200   \n\n   results[349].item_elo  \n0                   1200  \n\n[1 rows x 419 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>elo.scale</th>\n      <th>elo.prior_rating</th>\n      <th>elo.prior_weight</th>\n      <th>summary.accuracy</th>\n      <th>summary.weighted_points</th>\n      <th>summary.possible_weighted_points</th>\n      <th>summary.weighted_accuracy</th>\n      <th>summary.overall_elo</th>\n      <th>summary.overall_elo_unrounded</th>\n      <th>summary.categories.language_completion.accuracy</th>\n      <th>...</th>\n      <th>results[340].item_elo</th>\n      <th>results[341].item_elo</th>\n      <th>results[342].item_elo</th>\n      <th>results[343].item_elo</th>\n      <th>results[344].item_elo</th>\n      <th>results[345].item_elo</th>\n      <th>results[346].item_elo</th>\n      <th>results[347].item_elo</th>\n      <th>results[348].item_elo</th>\n      <th>results[349].item_elo</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>400.0</td>\n      <td>1000.0</td>\n      <td>4.0</td>\n      <td>0.534286</td>\n      <td>326.9</td>\n      <td>668.1875</td>\n      <td>0.489234</td>\n      <td>1007</td>\n      <td>1007.356924</td>\n      <td>0.9</td>\n      <td>...</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n      <td>1200</td>\n    </tr>\n  </tbody>\n</table>\n<p>1 rows × 419 columns</p>\n</div>"},"metadata":{}},{"name":"stdout","text":"\nCOMMAND: /usr/bin/python3 /root/.cache/huggingface/hub/datasets--AxiomicLabs--Arithmark-3.0/snapshots/6f6e59dd9b7e2c63455f7af7f838f9ecc3d0a746/bencharithmark-3.py --model /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model --device cuda --dtype float16 --batch-size 32 --max-context 2048 --data-path /root/.cache/huggingface/hub/datasets--AxiomicLabs--Arithmark-3.0/snapshots/6f6e59dd9b7e2c63455f7af7f838f9ecc3d0a746/arithmark-3.jsonl --primary-metric acc_norm --results-dir /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/arithmark3\nLoaded 1000 ArithMark 3.0 examples from /root/.cache/huggingface/hub/datasets--AxiomicLabs--Arithmark-3.0/blobs/237840149650455c4c54ff18f02203e2a76c1d75\nArithMark 3.0 dataset SHA-256: bf8ab1a5193d52cdf0e05ff0b3ca226bdfcf416cb6e75562dcbe72e7e4559435\nUsing device: cuda\n\n====================================================================\n  Loading /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model...\n====================================================================\n  90,148,352 parameters (torch.float16)\n\n  tokenizing:   0%|          | 0/1000 [00:00<?, ?it/s]\n  tokenizing:  15%|█▌        | 150/1000 [00:00<00:00, 1497.66it/s]\n  tokenizing:  30%|███       | 305/1000 [00:00<00:00, 1528.28it/s]\n  tokenizing:  46%|████▌     | 458/1000 [00:00<00:00, 1523.34it/s]\n  tokenizing:  61%|██████    | 611/1000 [00:00<00:00, 1515.49it/s]\n  tokenizing:  77%|███████▋  | 767/1000 [00:00<00:00, 1527.24it/s]\n  tokenizing:  92%|█████████▏| 921/1000 [00:00<00:00, 1530.70it/s]\n                                                                  \n\n  arithmark-3:   0%|          | 0/32 [00:00<?, ?it/s]\n  arithmark-3:   3%|▎         | 1/32 [00:00<00:21,  1.47it/s]\n  arithmark-3:   6%|▋         | 2/32 [00:00<00:12,  2.49it/s]\n  arithmark-3:   9%|▉         | 3/32 [00:01<00:09,  3.12it/s]\n  arithmark-3:  12%|█▎        | 4/32 [00:01<00:07,  3.54it/s]\n  arithmark-3:  16%|█▌        | 5/32 [00:01<00:07,  3.78it/s]\n  arithmark-3:  19%|█▉        | 6/32 [00:01<00:06,  3.95it/s]\n  arithmark-3:  22%|██▏       | 7/32 [00:02<00:06,  3.99it/s]\n  arithmark-3:  25%|██▌       | 8/32 [00:02<00:06,  3.98it/s]\n  arithmark-3:  28%|██▊       | 9/32 [00:02<00:05,  3.99it/s]\n  arithmark-3:  31%|███▏      | 10/32 [00:02<00:05,  3.96it/s]\n  arithmark-3:  34%|███▍      | 11/32 [00:03<00:05,  3.92it/s]\n  arithmark-3:  38%|███▊      | 12/32 [00:03<00:05,  3.87it/s]\n  arithmark-3:  41%|████      | 13/32 [00:03<00:04,  3.84it/s]\n  arithmark-3:  44%|████▍     | 14/32 [00:03<00:04,  3.79it/s]\n  arithmark-3:  47%|████▋     | 15/32 [00:04<00:04,  3.75it/s]\n  arithmark-3:  50%|█████     | 16/32 [00:04<00:04,  3.70it/s]\n  arithmark-3:  53%|█████▎    | 17/32 [00:04<00:04,  3.65it/s]\n  arithmark-3:  56%|█████▋    | 18/32 [00:04<00:03,  3.58it/s]\n  arithmark-3:  59%|█████▉    | 19/32 [00:05<00:03,  3.49it/s]\n  arithmark-3:  62%|██████▎   | 20/32 [00:05<00:03,  3.41it/s]\n  arithmark-3:  66%|██████▌   | 21/32 [00:05<00:03,  3.35it/s]\n  arithmark-3:  69%|██████▉   | 22/32 [00:06<00:03,  3.30it/s]\n  arithmark-3:  72%|███████▏  | 23/32 [00:06<00:02,  3.25it/s]\n  arithmark-3:  75%|███████▌  | 24/32 [00:06<00:02,  3.19it/s]\n  arithmark-3:  78%|███████▊  | 25/32 [00:07<00:02,  3.12it/s]\n  arithmark-3:  81%|████████▏ | 26/32 [00:07<00:01,  3.06it/s]\n  arithmark-3:  84%|████████▍ | 27/32 [00:07<00:01,  2.98it/s]\n  arithmark-3:  88%|████████▊ | 28/32 [00:08<00:01,  2.90it/s]\n  arithmark-3:  91%|█████████ | 29/32 [00:08<00:01,  2.76it/s]\n  arithmark-3:  94%|█████████▍| 30/32 [00:09<00:00,  2.58it/s]\n  arithmark-3:  97%|█████████▋| 31/32 [00:09<00:00,  2.43it/s]\n  arithmark-3: 100%|██████████| 32/32 [00:09<00:00,  3.05it/s]\n  arithmark-3: 100%|██████████| 32/32 [00:09<00:00,  3.29it/s]\n  arithmark-3: raw 44.90% (449/1000)  normalized 44.80% (448/1000)\n  speed: 102.8 examples/s  (tokenize 0.66s, evaluate 9.72s)\n\n  Category                                                             N       Raw   Normalized\n  ----------------------------------------------------------------------------------------------\n  elementary_school_math_continuation::addition::grades_1_2::easy    128    33.59%       33.59%\n  elementary_school_math_continuation::comparison::grades_2_3::medium    44    38.64%       38.64%\n  elementary_school_math_continuation::comparison_difference::grades_2_3::medium    48    39.58%       39.58%\n  elementary_school_math_continuation::data::grades_2_3::easy         43    34.88%       34.88%\n  elementary_school_math_continuation::division::grades_3_4::medium    54    48.15%       48.15%\n  elementary_school_math_continuation::fractions_counting::grades_3_4::medium    50    20.00%       18.00%\n  elementary_school_math_continuation::geometry_area::grades_4_5::medium    52    84.62%       86.54%\n  elementary_school_math_continuation::geometry_perimeter::grades_4_5::medium    45    66.67%       66.67%\n  elementary_school_math_continuation::measurement::grades_2_3::easy    76    36.84%       36.84%\n  elementary_school_math_continuation::money::grades_3_4::medium      64    35.94%       34.38%\n  elementary_school_math_continuation::multiplication::grades_3_4::medium    74    77.03%       75.68%\n  elementary_school_math_continuation::patterns::grades_3_4::medium    53    37.74%       37.74%\n  elementary_school_math_continuation::subtraction::grades_1_2::easy   117    30.77%       30.77%\n  elementary_school_math_continuation::time::grades_2_3::easy         55   100.00%      100.00%\n  elementary_school_math_continuation::two_step_add_subtract::grades_2_3::medium    46    26.09%       26.09%\n  elementary_school_math_continuation::two_step_addition::grades_2_3::medium    19    21.05%       21.05%\n  elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium    32    31.25%       34.38%\n\n====================================================================\n  /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/_void_model (90,148,352 params) RESULTS\n====================================================================\n  Raw continuation accuracy        44.90%\n  Length-normalized accuracy       44.80%\n  Primary (acc_norm)           44.80%\n====================================================================\nResults saved to /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/arithmark3/_kaggle_working_void_eval_appvoid--void-byte_92c601560cc2179e3bfc33e067fdd3bf9254e163__void_model_arithmark-3_results.json\nResult JSON: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/arithmark3/_kaggle_working_void_eval_appvoid--void-byte_92c601560cc2179e3bfc33e067fdd3bf9254e163__void_model_arithmark-3_results.json\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"   results.arithmark-3.acc  results.arithmark-3.acc_norm  \\\n0                     44.9                          44.8   \n\n   results.arithmark-3.raw_correct  results.arithmark-3.norm_correct  \\\n0                              449                               448   \n\n   results.arithmark-3.total  \\\n0                       1000   \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc  \\\n0                                           33.59375                                                    \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc_norm  \\\n0                                           33.59375                                                         \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.raw_correct  \\\n0                                                 43                                                            \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.norm_correct  \\\n0                                                 43                                                             \n\n   results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.total  \\\n0                                                128                                                      \n\n   ...  \\\n0  ...   \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.acc_norm  \\\n0                                          21.052632                                                                    \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.raw_correct  \\\n0                                                  4                                                                       \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.norm_correct  \\\n0                                                  4                                                                        \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.total  \\\n0                                                 19                                                                 \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc  \\\n0                                              31.25                                                                  \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc_norm  \\\n0                                             34.375                                                                       \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.raw_correct  \\\n0                                                 10                                                                          \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.norm_correct  \\\n0                                                 11                                                                           \n\n   results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.total  \\\n0                                                 32                                                                    \n\n   results.arithmark-3.primary_acc  \n0                             44.8  \n\n[1 rows x 91 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>results.arithmark-3.acc</th>\n      <th>results.arithmark-3.acc_norm</th>\n      <th>results.arithmark-3.raw_correct</th>\n      <th>results.arithmark-3.norm_correct</th>\n      <th>results.arithmark-3.total</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.acc_norm</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.raw_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.norm_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::addition::grades_1_2::easy.total</th>\n      <th>...</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.acc_norm</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.raw_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.norm_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_addition::grades_2_3::medium.total</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.acc_norm</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.raw_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.norm_correct</th>\n      <th>results.arithmark-3.categories.elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium.total</th>\n      <th>results.arithmark-3.primary_acc</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>44.9</td>\n      <td>44.8</td>\n      <td>449</td>\n      <td>448</td>\n      <td>1000</td>\n      <td>33.59375</td>\n      <td>33.59375</td>\n      <td>43</td>\n      <td>43</td>\n      <td>128</td>\n      <td>...</td>\n      <td>21.052632</td>\n      <td>4</td>\n      <td>4</td>\n      <td>19</td>\n      <td>31.25</td>\n      <td>34.375</td>\n      <td>10</td>\n      <td>11</td>\n      <td>32</td>\n      <td>44.8</td>\n    </tr>\n  </tbody>\n</table>\n<p>1 rows × 91 columns</p>\n</div>"},"metadata":{}},{"name":"stdout","text":"\nUnified summary\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"  model           model_id                                  revision  \\\n0  Void  appvoid/void-byte  92c601560cc2179e3bfc33e067fdd3bf9254e163   \n\n   hellaswag       piqa   arc_easy  arc_challenge  arithmark3  bananamind_elo  \\\n0  38.677554  67.464635  47.306397      28.156997        44.8          1007.0   \n\n   bananamind_accuracy  bananamind_weighted_accuracy  intelligence_index  \\\n0             0.534286                      0.489234           23.918248   \n\n   benchmark_max_length  \n0                  2048  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>model</th>\n      <th>model_id</th>\n      <th>revision</th>\n      <th>hellaswag</th>\n      <th>piqa</th>\n      <th>arc_easy</th>\n      <th>arc_challenge</th>\n      <th>arithmark3</th>\n      <th>bananamind_elo</th>\n      <th>bananamind_accuracy</th>\n      <th>bananamind_weighted_accuracy</th>\n      <th>intelligence_index</th>\n      <th>benchmark_max_length</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Void</td>\n      <td>appvoid/void-byte</td>\n      <td>92c601560cc2179e3bfc33e067fdd3bf9254e163</td>\n      <td>38.677554</td>\n      <td>67.464635</td>\n      <td>47.306397</td>\n      <td>28.156997</td>\n      <td>44.8</td>\n      <td>1007.0</td>\n      <td>0.534286</td>\n      <td>0.489234</td>\n      <td>23.918248</td>\n      <td>2048</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}},{"name":"stdout","text":"Saved: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163/void_unified_summary.json\n","output_type":"stream"}],"execution_count":19},{"id":"39ab7ade","cell_type":"markdown","source":"## 11. Plot benchmark scores","metadata":{"id":"39ab7ade"}},{"id":"be3ddd24","cell_type":"code","source":"import matplotlib.pyplot as plt\n\ndef plot_void_scores(\n    row,\n    columns=None,\n    title=\"Void benchmark performance\",\n):\n    if isinstance(row, dict):\n        df = pd.DataFrame([row])\n    else:\n        df = row.copy()\n\n    if columns is None:\n        columns = [\n            c for c in [\n                \"hellaswag\",\n                \"piqa\",\n                \"arc_easy\",\n                \"arc_challenge\",\n                \"arithmark3\",\n                \"intelligence_index\",\n            ]\n            if c in df.columns\n        ]\n\n    if not columns:\n        raise ValueError(\"No benchmark score columns found.\")\n\n    values = df.iloc[0][columns].astype(float)\n    ax = values.plot(kind=\"bar\", figsize=(10, 5))\n    ax.set_title(title)\n    ax.set_ylabel(\"Score\")\n    ax.set_xlabel(\"Benchmark\")\n    ax.grid(axis=\"y\", alpha=0.25)\n    plt.tight_layout()\n    plt.show()\n\nplot_void_scores(void_scores)\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":395},"id":"be3ddd24","outputId":"ce2e0597-e679-452d-c51a-cfaffef0956e","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:27:54.521200Z","iopub.execute_input":"2026-10-09T02:27:54.521634Z","iopub.status.idle":"2026-10-09T02:27:54.766658Z","shell.execute_reply.started":"2026-10-09T02:27:54.521605Z","shell.execute_reply":"2026-10-09T02:27:54.765819Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x500 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":20},{"id":"d89e7deb","cell_type":"markdown","source":"## 12. Package results","metadata":{"id":"d89e7deb"}},{"id":"7084e3be","cell_type":"code","source":"def package_results():\n    archive = shutil.make_archive(\n        str(ROOT),\n        \"zip\",\n        root_dir=str(ROOT.parent),\n        base_dir=ROOT.name,\n    )\n    print(\"Created:\", archive)\n    return archive\n\nresults_zip = package_results()\n","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7084e3be","outputId":"e0bbd15e-a5d5-4fc4-cb3a-e13b5710883a","trusted":true,"jupyter":{"source_hidden":true},"execution":{"iopub.status.busy":"2026-10-09T02:27:54.767591Z","iopub.execute_input":"2026-10-09T02:27:54.767843Z","iopub.status.idle":"2026-10-09T02:28:13.087976Z","shell.execute_reply.started":"2026-10-09T02:27:54.767820Z","shell.execute_reply":"2026-10-09T02:28:13.087231Z"}},"outputs":[{"name":"stdout","text":"Created: /kaggle/working/void_eval/appvoid--void-byte/92c601560cc2179e3bfc33e067fdd3bf9254e163.zip\n","output_type":"stream"}],"execution_count":21},{"id":"18dbbb7c","cell_type":"markdown","source":"## 13. Suggested workflow\n\n### A. Verify the pinned Void checkpoint\n\n```python\nprint(ACTIVE_MODEL)\nprint(\"revision:\", RESOLVED_MODEL_REVISION)\nprint(\"params:\", sum(p.numel() for p in void_model.parameters()))\nprint(\"context:\", void_model.config.max_position_embeddings)\nprint(\"layers:\", void_model.config.num_hidden_layers)\n```\n\nExpected architecture from the supplied training notebook:\n\n```text\n90,148,352 parameters\n24 layers\nhidden size 640\nMLP 1440\n10 query heads / 2 KV heads\nhead_dim 64\nvocab 259\nnative context 2048\n```\n\n### B. Test generation\n\n```python\ngenerate_void(\"The capital of France is\", max_new_bytes=64)\n```\n\n### C. Smoke-test the byte-level path\n\n```python\npiqa_smoke = run_open_slm(tasks=[\"piqa\"], limit=50)\n```\n\n### D. Run the same four Open SLM tasks used for Fig\n\n```python\nopen_scores = run_open_slm()\n```\n\n### E. Run the complete benchmark suite\n\n```python\nvoid_scores = run_full_suite()\n```\n\n### F. Package all results\n\n```python\nresults_zip = package_results()\n```\n\n### Native-context variant\n\nThe default `BENCH_MAX_LENGTH = 1024` intentionally matches the earlier Fig benchmark. To measure Void with its full trained context instead, set:\n\n```python\nBENCH_MAX_LENGTH = 2048\nLM_EVAL_MAX_LENGTH = BENCH_MAX_LENGTH\n```\n\nthen rerun the benchmark cells. Keep the 1024 run if you want the cleanest apples-to-apples comparison against Fig.\n","metadata":{"id":"18dbbb7c"}}]}