Add

.gitattributes

@@ -32,3 +32,27 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_0.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_4.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-web_nlg_en_PALM_prompt_0.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-web_nlg_en_PALM_prompt_4.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-web_nlg_en_PALM_prompt_5.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_2.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_5.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_2.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_3.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-web_nlg_en_PALM_prompt_1.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-web_nlg_en_PALM_prompt_3.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_0.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_1.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_3.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_4.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_0.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_2.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_3.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_5.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_5.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_1.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-web_nlg_en_PALM_prompt_2.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_1.jsonl filter=lfs diff=lfs merge=lfs -text
+evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_4.jsonl filter=lfs diff=lfs merge=lfs -text

evaluation/generation/agg.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_0.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "bleu": 3.9626050306721474, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.06024253979194259}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_precision": 0.1540985111881967, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0014515620097711474}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_recall": 0.31140262883046665, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.0024251988602698067}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_fmeasure": 0.20284230759257002, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0017464348006575089}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_precision": 0.05355018382472272, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.0009106172876128309}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_recall": 0.10690154622755875, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0017688696499961142}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_fmeasure": 0.07031535968691954, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.001168196298024764}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_precision": 0.1327617315507999, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.001197611976138011}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_recall": 0.2711114650902915, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0021100724050297824}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_fmeasure": 0.1753585977687643, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0014608296346256319}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_precision": 0.1387916405736721, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.0013406445609266567}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_recall": 0.2810609872616038, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.0022804790016154107}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_fmeasure": 0.18282034804175978, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0016241979991028408}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 0, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_1.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "bleu": 11.593827123577368, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.12407021076125696}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_precision": 0.5509122991009537, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0033661250504397264}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_recall": 0.42140078880413323, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.002953747532093766}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_fmeasure": 0.45020216196031243, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0023747076071133698}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_precision": 0.2650742428626904, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.0027402930606959166}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_recall": 0.19921132653073137, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0021274437677356846}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_fmeasure": 0.21317325143407617, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.0020127889592610262}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_precision": 0.40311524575775876, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.003054424817440509}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_recall": 0.30479322858663915, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0023980133198678024}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_fmeasure": 0.32668549807617586, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.002086318029186934}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_precision": 0.4523945212247963, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.00326815326036064}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_recall": 0.3446091324717908, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.0027191831013193444}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_fmeasure": 0.3686111369512851, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0023418416254223887}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 1, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_2.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "bleu": 12.88190096920724, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.17583922253090617}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_precision": 0.5822720419906148, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0032962978807536585}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_recall": 0.4337280880183242, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.0028431977686949614}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_fmeasure": 0.4721264666575481, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.002268547262390229}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_precision": 0.2931157748619445, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.0028165108542885105}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_recall": 0.2147077046167617, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0021786795843148085}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_fmeasure": 0.23375071908985015, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.0020523859536162883}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_precision": 0.4369685674136242, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.003073517456588823}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_recall": 0.32297689774894583, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0024097063140581243}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_fmeasure": 0.3522716641150115, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0020991364053741914}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_precision": 0.4863687819740207, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.0032507530347318466}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_recall": 0.3613566974986443, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.0026568660208198115}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_fmeasure": 0.393682260751447, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0023007053568388975}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 2, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_3.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "bleu": 13.262436270778784, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.12966668718803287}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_precision": 0.5813922477147344, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0032594641701596022}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_recall": 0.43750260743630337, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.002850951212768957}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_fmeasure": 0.47525930406845607, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0022663496963600087}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_precision": 0.2963848070406136, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.002811561987291426}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_recall": 0.2198532494536229, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0022399911777690423}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_fmeasure": 0.238646665315093, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.0020938647359093243}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_precision": 0.43608943087396823, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.0030728003407875415}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_recall": 0.32653357863256616, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.002478214580278462}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_fmeasure": 0.3550920935079866, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.002168420115905195}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_precision": 0.4873959951365985, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.0032309608357424142}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_recall": 0.3665706405925854, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.0027071065833861527}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_fmeasure": 0.3982122870175536, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.002333698912202139}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 3, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_4.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "bleu": 13.476662518674425, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.14804924963101374}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_precision": 0.575148309756594, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0032456743499240143}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_recall": 0.4379553251280355, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.0027875515895071923}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_fmeasure": 0.4751483029381261, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0022939871226639315}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_precision": 0.2932213385624628, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.0027850927611603377}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_recall": 0.220787156975807, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0022308643708334754}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_fmeasure": 0.23937430569965518, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.0021223157395609733}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_precision": 0.4297630033921168, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.0030250335936550877}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_recall": 0.3260216890650744, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0024522487516979575}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_fmeasure": 0.35385483221937813, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0021780339813240314}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_precision": 0.4813876805218902, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.0032181540128894265}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_recall": 0.36675888786779215, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.002699461696539119}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_fmeasure": 0.3976384682503085, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0023729927223769785}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 4, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_5.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "bleu": 13.156897203455948, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.08630136341873869}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_precision": 0.5738627336760422, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0032894505336520264}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_recall": 0.4353618479598648, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.0027767731270488508}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge1_fmeasure": 0.4729751649957838, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0022801925622949843}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_precision": 0.29086794940233046, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.002818639353471394}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_recall": 0.21741764823749096, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.002221377915423391}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rouge2_fmeasure": 0.23628094628078694, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.00211214055033469}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_precision": 0.4304412705229436, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.003073791049621951}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_recall": 0.32508975084283026, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0024528402950547467}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeL_fmeasure": 0.3534318578391924, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0021797527155027353}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_precision": 0.4815893698288744, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.0032662227256431515}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_recall": 0.36475434401700674, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.00266853275926591}, {"task_name": "e2e_nlg_cleaned", "prompt_name": "generate_text_restaurant", "rougeLsum_fmeasure": 0.39642244501779056, "fixed_answer_choice_list": null, "dataset_path": "e2e_nlg_cleaned", "dataset_name": null, "subset": null, "prompt_id": "1acabbc3-c9b9-4624-a684-29faeccff46f", "prompt_jinja": "Given the following data about a restaurant:\n{% for feature in meaning_representation.split(\"]\") %} {% set key = feature.split(\"[\")[0].replace(\",\",\"\") %} {% set value = feature.replace(\",\",\"\").replace(key+\"[\", '''') %}\n{% if value != \"\" %} {{key}} : {{value}} {% endif %}\n{%- endfor %}\nGenerate some text about this restaurant. ||| {{human_reference}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.002354679808047655}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 5, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_0.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_precision": 0.10711262762227787, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0020601915536585967}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_recall": 0.25337181483295146, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.0046856436696164905}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_fmeasure": 0.1478842341216362, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0027472203051295764}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_precision": 0.019781013833534398, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.0008722452654034713}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_recall": 0.04912277748197487, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0021576559260432712}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_fmeasure": 0.02780714664643612, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.0012134748561574466}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_precision": 0.08153762034631415, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.0015205040880351287}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_recall": 0.19377740656973594, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0035212605780510554}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_fmeasure": 0.11266163818049864, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0020299599299667603}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_precision": 0.08657778420332544, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.0016263154802718866}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_recall": 0.20583552514369016, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.003771689144281502}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_fmeasure": 0.11968068511417429, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0021748845339722685}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "bleu": 1.0285249742430023, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.04658561770880105}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 0, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_1.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_precision": 0.2154872949099143, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0037469904189591395}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_recall": 0.25867586586469277, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.004134512250176692}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_fmeasure": 0.2183934186796248, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0031131023395006203}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_precision": 0.0447245429308555, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.0020714108653434506}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_recall": 0.05502827731187604, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0023758194054523916}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_fmeasure": 0.04500583650450346, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.001863892149786922}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_precision": 0.1614628963956234, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.0030120325913654804}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_recall": 0.19472956411902964, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.0033235017592203077}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_fmeasure": 0.1635144970491249, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0024820064917382306}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_precision": 0.16439844895006708, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.003021699034691555}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_recall": 0.2002131667137321, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.003522860713518912}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_fmeasure": 0.1671666603817299, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0025413985087851003}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "bleu": 2.1792875082477123, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.06319420014636688}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 1, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_2.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_precision": 0.24578221078853943, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0038311545433087246}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_recall": 0.25247232169607386, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.003851791759690215}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_fmeasure": 0.23579050025596546, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0032208877086972645}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_precision": 0.05385976155670831, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.002293715583911734}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_recall": 0.05655894066313464, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.002406489537857907}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_fmeasure": 0.051882326943507785, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.002120030804437139}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_precision": 0.18601506569279253, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.0031620641585856746}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_recall": 0.19159840411716447, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.003222530909275831}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_fmeasure": 0.17839101172174274, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0026918945920387577}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_precision": 0.18783807621807994, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.003156835477899629}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_recall": 0.19452321319767416, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.0033253733370942227}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_fmeasure": 0.18050054360244336, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.002709070917463213}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "bleu": 2.9719119603418074, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.16832095088560176}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 2, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/agg.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_3.json

@@ -0,0 +1 @@
+{"results": [{"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_precision": 0.24477878269919434, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_precision_stderr": 0.0043304256076557015}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_recall": 0.23546035253406764, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_recall_stderr": 0.0040420090423460295}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge1_fmeasure": 0.2277409702430376, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge1_fmeasure_stderr": 0.0036665816219810165}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_precision": 0.05535540959023755, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_precision_stderr": 0.002520852290920342}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_recall": 0.05400488265902618, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_recall_stderr": 0.0023942103295597683}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rouge2_fmeasure": 0.05187808695380671, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rouge2_fmeasure_stderr": 0.0022736997047107283}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_precision": 0.18676731153569484, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_precision_stderr": 0.0035994106495257387}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_recall": 0.17809432971388495, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_recall_stderr": 0.003241778335625792}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeL_fmeasure": 0.1726765296689622, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeL_fmeasure_stderr": 0.0030037675906149446}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_precision": 0.1881533844870454, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_precision_stderr": 0.003601435982381053}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_recall": 0.18040945162861388, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_recall_stderr": 0.0033378628217040685}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "rougeLsum_fmeasure": 0.1742948037879305, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "rougeLsum_fmeasure_stderr": 0.0030227160469161974}, {"task_name": "gem_xsum", "prompt_name": "article_DOC_summary", "bleu": 3.2041870719788714, "fixed_answer_choice_list": null, "dataset_path": "GEM/xsum", "dataset_name": null, "subset": "", "prompt_id": "a8d4ecfa-c944-44d5-878c-04fd5db59e64", "prompt_jinja": "Article: {{document}}\n\nSummary: ||| {{target}}", "prompt_original_task": true, "comment": "", "bleu_stderr": 0.20611661341919532}], "config": {"model": "hf-causal", "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", "task_args": "", "num_fewshot": 3, "batch_size": 16, "device": "cuda", "use_cache": false, "limit": 3000, "bootstrap_iters": 10, "seed": 1234}}

evaluation/generation/examples.lm1-2b8-55b-oscarroots_GEM-wiki_lingua_en_tldr_en_0.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc909f9f5301033e82365578e8b8ae19e71f21eb19b609a2d9224ef10b7dba9a
+size 7681791

evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_0.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03a704484c627b14ffb0fcc4f3312ccd6beb4380ff0d76185385316ee5e8dbd3
+size 4469028

evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_1.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb122f106ee8bf88bd465ba736a7900c7f385c34b9565bc4b482a980f2ae76e0
+size 5011151

evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_2.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a236e3f38e5ce8163fd9b2ff5cc95e199fec013fb2e8731b5b09affbf9a4c991
+size 6079576

evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_3.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c5371e37f524bb77793a465c589433003aecaa7e13001172e07034d47de842c
+size 7163221

evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_4.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a1e03d6c709c1a3bbf2aec2f5920402630be4f5ce34bc7b0c4907c37c3e0556
+size 8246022

evaluation/generation/examples.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_5.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0a8dec4db5b0f7393c5775ee630799993a8e690b81844c1cc13ea9a1e4636d9
+size 9333092

evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_0.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a54ee931d6dc5464885b20e409838b7d09556b2c2ef2fd2721e8e857cc757cf0
+size 2835274

evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_1.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9168ce4ffb1a34cdc8b518867ec887d2bcfc6ff954189b0b47c4974287f318c
+size 4950440

evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_2.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2658ed5ea2dcb3ae823c8694147c2bde298e073e2d53d7674cc17c1a91adaff
+size 7189518

evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_3.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d52704d2c1ca8ce3bdc30e0a89559b744b14ff1edfcfdd104e76f39f2caf4376
+size 9463905

evaluation/generation/examples.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_4.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17d4acf50db6ce7936d568d2e3164ed9fdea179db9416d5b5a5be518e211782b
+size 11627724

evaluation/generation/slim.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_0.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "bleu": 3.9626050306721474,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "bleu_stderr": 0.06024253979194259
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_precision": 0.1540985111881967,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_precision_stderr": 0.0014515620097711474
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_recall": 0.31140262883046665,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_recall_stderr": 0.0024251988602698067
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_fmeasure": 0.20284230759257002,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_fmeasure_stderr": 0.0017464348006575089
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_precision": 0.05355018382472272,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_precision_stderr": 0.0009106172876128309
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_recall": 0.10690154622755875,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_recall_stderr": 0.0017688696499961142
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_fmeasure": 0.07031535968691954,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_fmeasure_stderr": 0.001168196298024764
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_precision": 0.1327617315507999,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_precision_stderr": 0.001197611976138011
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_recall": 0.2711114650902915,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_recall_stderr": 0.0021100724050297824
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_fmeasure": 0.1753585977687643,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_fmeasure_stderr": 0.0014608296346256319
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_precision": 0.1387916405736721,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_precision_stderr": 0.0013406445609266567
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_recall": 0.2810609872616038,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_recall_stderr": 0.0022804790016154107
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_fmeasure": 0.18282034804175978,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_fmeasure_stderr": 0.0016241979991028408
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 0,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_1.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "bleu": 11.593827123577368,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "bleu_stderr": 0.12407021076125696
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_precision": 0.5509122991009537,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_precision_stderr": 0.0033661250504397264
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_recall": 0.42140078880413323,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_recall_stderr": 0.002953747532093766
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_fmeasure": 0.45020216196031243,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_fmeasure_stderr": 0.0023747076071133698
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_precision": 0.2650742428626904,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_precision_stderr": 0.0027402930606959166
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_recall": 0.19921132653073137,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_recall_stderr": 0.0021274437677356846
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_fmeasure": 0.21317325143407617,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_fmeasure_stderr": 0.0020127889592610262
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_precision": 0.40311524575775876,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_precision_stderr": 0.003054424817440509
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_recall": 0.30479322858663915,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_recall_stderr": 0.0023980133198678024
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_fmeasure": 0.32668549807617586,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_fmeasure_stderr": 0.002086318029186934
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_precision": 0.4523945212247963,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_precision_stderr": 0.00326815326036064
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_recall": 0.3446091324717908,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_recall_stderr": 0.0027191831013193444
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_fmeasure": 0.3686111369512851,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_fmeasure_stderr": 0.0023418416254223887
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 1,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_2.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "bleu": 12.88190096920724,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "bleu_stderr": 0.17583922253090617
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_precision": 0.5822720419906148,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_precision_stderr": 0.0032962978807536585
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_recall": 0.4337280880183242,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_recall_stderr": 0.0028431977686949614
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_fmeasure": 0.4721264666575481,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_fmeasure_stderr": 0.002268547262390229
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_precision": 0.2931157748619445,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_precision_stderr": 0.0028165108542885105
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_recall": 0.2147077046167617,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_recall_stderr": 0.0021786795843148085
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_fmeasure": 0.23375071908985015,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_fmeasure_stderr": 0.0020523859536162883
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_precision": 0.4369685674136242,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_precision_stderr": 0.003073517456588823
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_recall": 0.32297689774894583,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_recall_stderr": 0.0024097063140581243
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_fmeasure": 0.3522716641150115,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_fmeasure_stderr": 0.0020991364053741914
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_precision": 0.4863687819740207,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_precision_stderr": 0.0032507530347318466
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_recall": 0.3613566974986443,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_recall_stderr": 0.0026568660208198115
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_fmeasure": 0.393682260751447,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_fmeasure_stderr": 0.0023007053568388975
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 2,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_3.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "bleu": 13.262436270778784,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "bleu_stderr": 0.12966668718803287
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_precision": 0.5813922477147344,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_precision_stderr": 0.0032594641701596022
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_recall": 0.43750260743630337,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_recall_stderr": 0.002850951212768957
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_fmeasure": 0.47525930406845607,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_fmeasure_stderr": 0.0022663496963600087
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_precision": 0.2963848070406136,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_precision_stderr": 0.002811561987291426
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_recall": 0.2198532494536229,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_recall_stderr": 0.0022399911777690423
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_fmeasure": 0.238646665315093,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_fmeasure_stderr": 0.0020938647359093243
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_precision": 0.43608943087396823,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_precision_stderr": 0.0030728003407875415
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_recall": 0.32653357863256616,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_recall_stderr": 0.002478214580278462
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_fmeasure": 0.3550920935079866,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_fmeasure_stderr": 0.002168420115905195
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_precision": 0.4873959951365985,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_precision_stderr": 0.0032309608357424142
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_recall": 0.3665706405925854,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_recall_stderr": 0.0027071065833861527
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_fmeasure": 0.3982122870175536,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_fmeasure_stderr": 0.002333698912202139
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 3,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_4.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "bleu": 13.476662518674425,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "bleu_stderr": 0.14804924963101374
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_precision": 0.575148309756594,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_precision_stderr": 0.0032456743499240143
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_recall": 0.4379553251280355,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_recall_stderr": 0.0027875515895071923
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_fmeasure": 0.4751483029381261,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_fmeasure_stderr": 0.0022939871226639315
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_precision": 0.2932213385624628,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_precision_stderr": 0.0027850927611603377
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_recall": 0.220787156975807,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_recall_stderr": 0.0022308643708334754
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_fmeasure": 0.23937430569965518,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_fmeasure_stderr": 0.0021223157395609733
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_precision": 0.4297630033921168,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_precision_stderr": 0.0030250335936550877
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_recall": 0.3260216890650744,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_recall_stderr": 0.0024522487516979575
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_fmeasure": 0.35385483221937813,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_fmeasure_stderr": 0.0021780339813240314
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_precision": 0.4813876805218902,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_precision_stderr": 0.0032181540128894265
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_recall": 0.36675888786779215,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_recall_stderr": 0.002699461696539119
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_fmeasure": 0.3976384682503085,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_fmeasure_stderr": 0.0023729927223769785
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 4,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_e2e_nlg_cleaned_generate_text_restaurant_5.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "bleu": 13.156897203455948,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "bleu_stderr": 0.08630136341873869
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_precision": 0.5738627336760422,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_precision_stderr": 0.0032894505336520264
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_recall": 0.4353618479598648,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_recall_stderr": 0.0027767731270488508
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge1_fmeasure": 0.4729751649957838,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge1_fmeasure_stderr": 0.0022801925622949843
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_precision": 0.29086794940233046,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_precision_stderr": 0.002818639353471394
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_recall": 0.21741764823749096,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_recall_stderr": 0.002221377915423391
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rouge2_fmeasure": 0.23628094628078694,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rouge2_fmeasure_stderr": 0.00211214055033469
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_precision": 0.4304412705229436,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_precision_stderr": 0.003073791049621951
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_recall": 0.32508975084283026,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_recall_stderr": 0.0024528402950547467
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeL_fmeasure": 0.3534318578391924,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeL_fmeasure_stderr": 0.0021797527155027353
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_precision": 0.4815893698288744,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_precision_stderr": 0.0032662227256431515
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_recall": 0.36475434401700674,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_recall_stderr": 0.00266853275926591
+    },
+    {
+      "task_name": "e2e_nlg_cleaned",
+      "prompt_name": "generate_text_restaurant",
+      "rougeLsum_fmeasure": 0.39642244501779056,
+      "dataset_path": "e2e_nlg_cleaned",
+      "dataset_name": null,
+      "subset": null,
+      "rougeLsum_fmeasure_stderr": 0.002354679808047655
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 5,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_0.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_precision": 0.10711262762227787,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_precision_stderr": 0.0020601915536585967
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_recall": 0.25337181483295146,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_recall_stderr": 0.0046856436696164905
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_fmeasure": 0.1478842341216362,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_fmeasure_stderr": 0.0027472203051295764
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_precision": 0.019781013833534398,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_precision_stderr": 0.0008722452654034713
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_recall": 0.04912277748197487,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_recall_stderr": 0.0021576559260432712
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_fmeasure": 0.02780714664643612,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_fmeasure_stderr": 0.0012134748561574466
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_precision": 0.08153762034631415,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_precision_stderr": 0.0015205040880351287
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_recall": 0.19377740656973594,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_recall_stderr": 0.0035212605780510554
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_fmeasure": 0.11266163818049864,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_fmeasure_stderr": 0.0020299599299667603
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_precision": 0.08657778420332544,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_precision_stderr": 0.0016263154802718866
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_recall": 0.20583552514369016,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_recall_stderr": 0.003771689144281502
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_fmeasure": 0.11968068511417429,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_fmeasure_stderr": 0.0021748845339722685
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "bleu": 1.0285249742430023,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "bleu_stderr": 0.04658561770880105
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 0,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_1.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_precision": 0.2154872949099143,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_precision_stderr": 0.0037469904189591395
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_recall": 0.25867586586469277,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_recall_stderr": 0.004134512250176692
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_fmeasure": 0.2183934186796248,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_fmeasure_stderr": 0.0031131023395006203
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_precision": 0.0447245429308555,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_precision_stderr": 0.0020714108653434506
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_recall": 0.05502827731187604,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_recall_stderr": 0.0023758194054523916
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_fmeasure": 0.04500583650450346,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_fmeasure_stderr": 0.001863892149786922
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_precision": 0.1614628963956234,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_precision_stderr": 0.0030120325913654804
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_recall": 0.19472956411902964,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_recall_stderr": 0.0033235017592203077
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_fmeasure": 0.1635144970491249,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_fmeasure_stderr": 0.0024820064917382306
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_precision": 0.16439844895006708,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_precision_stderr": 0.003021699034691555
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_recall": 0.2002131667137321,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_recall_stderr": 0.003522860713518912
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_fmeasure": 0.1671666603817299,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_fmeasure_stderr": 0.0025413985087851003
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "bleu": 2.1792875082477123,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "bleu_stderr": 0.06319420014636688
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 1,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_2.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_precision": 0.24578221078853943,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_precision_stderr": 0.0038311545433087246
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_recall": 0.25247232169607386,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_recall_stderr": 0.003851791759690215
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_fmeasure": 0.23579050025596546,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_fmeasure_stderr": 0.0032208877086972645
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_precision": 0.05385976155670831,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_precision_stderr": 0.002293715583911734
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_recall": 0.05655894066313464,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_recall_stderr": 0.002406489537857907
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_fmeasure": 0.051882326943507785,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_fmeasure_stderr": 0.002120030804437139
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_precision": 0.18601506569279253,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_precision_stderr": 0.0031620641585856746
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_recall": 0.19159840411716447,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_recall_stderr": 0.003222530909275831
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_fmeasure": 0.17839101172174274,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_fmeasure_stderr": 0.0026918945920387577
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_precision": 0.18783807621807994,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_precision_stderr": 0.003156835477899629
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_recall": 0.19452321319767416,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_recall_stderr": 0.0033253733370942227
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_fmeasure": 0.18050054360244336,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_fmeasure_stderr": 0.002709070917463213
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "bleu": 2.9719119603418074,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "bleu_stderr": 0.16832095088560176
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 2,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/generation/slim.lm1-2b8-55b-oscarroots_gem_xsum_article_DOC_summary_3.json

@@ -0,0 +1,133 @@
+{
+  "results": [
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_precision": 0.24477878269919434,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_precision_stderr": 0.0043304256076557015
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_recall": 0.23546035253406764,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_recall_stderr": 0.0040420090423460295
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge1_fmeasure": 0.2277409702430376,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge1_fmeasure_stderr": 0.0036665816219810165
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_precision": 0.05535540959023755,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_precision_stderr": 0.002520852290920342
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_recall": 0.05400488265902618,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_recall_stderr": 0.0023942103295597683
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rouge2_fmeasure": 0.05187808695380671,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rouge2_fmeasure_stderr": 0.0022736997047107283
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_precision": 0.18676731153569484,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_precision_stderr": 0.0035994106495257387
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_recall": 0.17809432971388495,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_recall_stderr": 0.003241778335625792
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeL_fmeasure": 0.1726765296689622,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeL_fmeasure_stderr": 0.0030037675906149446
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_precision": 0.1881533844870454,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_precision_stderr": 0.003601435982381053
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_recall": 0.18040945162861388,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_recall_stderr": 0.0033378628217040685
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "rougeLsum_fmeasure": 0.1742948037879305,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "rougeLsum_fmeasure_stderr": 0.0030227160469161974
+    },
+    {
+      "task_name": "gem_xsum",
+      "prompt_name": "article_DOC_summary",
+      "bleu": 3.2041870719788714,
+      "dataset_path": "GEM/xsum",
+      "dataset_name": null,
+      "subset": "",
+      "bleu_stderr": 0.20611661341919532
+    }
+  ],
+  "config": {
+    "model": "hf-causal",
+    "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-oscarroots/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
+    "task_args": "",
+    "num_fewshot": 3,
+    "batch_size": 16,
+    "device": "cuda",
+    "use_cache": false,
+    "limit": 3000,
+    "bootstrap_iters": 10,
+    "seed": 1234
+  }
+}

evaluation/rankeval/lm1-2b8-55b-oscarroots_0.json

@@ -0,0 +1,32 @@
+{
+    "results": {
+        "anli_r1": {
+            "acc": 0.322,
+            "acc_stderr": 0.014782913600996673
+        },
+        "anli_r2": {
+            "acc": 0.331,
+            "acc_stderr": 0.014888272588203936
+        },
+        "anli_r3": {
+            "acc": 0.3358333333333333,
+            "acc_stderr": 0.013639261190932887
+        },
+        "cb": {
+            "acc": 0.35714285714285715,
+            "acc_stderr": 0.0646095738380922,
+            "f1": 0.24888576120103215
+        },
+        "copa": {
+            "acc": 0.68,
+            "acc_stderr": 0.046882617226215034
+        }
+    },
+    "versions": {
+        "anli_r1": 0,
+        "anli_r2": 0,
+        "anli_r3": 0,
+        "cb": 1,
+        "copa": 0
+    }
+}

evaluation/rankeval/lm1-2b8-55b-oscarroots_0_lm-eval_global_step52452_2023-02-25-11-16-27_0shots_backup.json

@@ -0,0 +1,32 @@
+{
+    "results": {
+        "anli_r1": {
+            "acc": 0.322,
+            "acc_stderr": 0.014782913600996673
+        },
+        "anli_r2": {
+            "acc": 0.331,
+            "acc_stderr": 0.014888272588203936
+        },
+        "anli_r3": {
+            "acc": 0.3358333333333333,
+            "acc_stderr": 0.013639261190932887
+        },
+        "cb": {
+            "acc": 0.35714285714285715,
+            "acc_stderr": 0.0646095738380922,
+            "f1": 0.24888576120103215
+        },
+        "copa": {
+            "acc": 0.68,
+            "acc_stderr": 0.046882617226215034
+        }
+    },
+    "versions": {
+        "anli_r1": 0,
+        "anli_r2": 0,
+        "anli_r3": 0,
+        "cb": 1,
+        "copa": 0
+    }
+}

evaluation/rankeval/lm1-2b8-55b-oscarroots_1.json

@@ -0,0 +1,32 @@
+{
+    "results": {
+        "anli_r1": {
+            "acc": 0.303,
+            "acc_stderr": 0.014539683710535255
+        },
+        "anli_r2": {
+            "acc": 0.319,
+            "acc_stderr": 0.014746404865473486
+        },
+        "anli_r3": {
+            "acc": 0.33416666666666667,
+            "acc_stderr": 0.013622434813136781
+        },
+        "cb": {
+            "acc": 0.48214285714285715,
+            "acc_stderr": 0.06737697508644648,
+            "f1": 0.3270348837209302
+        },
+        "copa": {
+            "acc": 0.66,
+            "acc_stderr": 0.04760952285695237
+        }
+    },
+    "versions": {
+        "anli_r1": 0,
+        "anli_r2": 0,
+        "anli_r3": 0,
+        "cb": 1,
+        "copa": 0
+    }
+}

evaluation/rankeval/lm1-2b8-55b-oscarroots_1_lm-eval_global_step52452_2023-02-25-11-18-29_1shots_backup.json

@@ -0,0 +1,32 @@
+{
+    "results": {
+        "anli_r1": {
+            "acc": 0.303,
+            "acc_stderr": 0.014539683710535255
+        },
+        "anli_r2": {
+            "acc": 0.319,
+            "acc_stderr": 0.014746404865473486
+        },
+        "anli_r3": {
+            "acc": 0.33416666666666667,
+            "acc_stderr": 0.013622434813136781
+        },
+        "cb": {
+            "acc": 0.48214285714285715,
+            "acc_stderr": 0.06737697508644648,
+            "f1": 0.3270348837209302
+        },
+        "copa": {
+            "acc": 0.66,
+            "acc_stderr": 0.04760952285695237
+        }
+    },
+    "versions": {
+        "anli_r1": 0,
+        "anli_r2": 0,
+        "anli_r3": 0,
+        "cb": 1,
+        "copa": 0
+    }
+}

evaluation/rankeval/lm1-2b8-55b-oscarroots_2.json

@@ -0,0 +1,32 @@
+{
+    "results": {
+        "anli_r1": {
+            "acc": 0.311,
+            "acc_stderr": 0.014645596385722694
+        },
+        "anli_r2": {
+            "acc": 0.346,
+            "acc_stderr": 0.01505026612756444
+        },
+        "anli_r3": {
+            "acc": 0.325,
+            "acc_stderr": 0.013526454480351014
+        },
+        "cb": {
+            "acc": 0.4107142857142857,
+            "acc_stderr": 0.0663363415035954,
+            "f1": 0.26927814732692784
+        },
+        "copa": {
+            "acc": 0.68,
+            "acc_stderr": 0.04688261722621505
+        }
+    },
+    "versions": {
+        "anli_r1": 0,
+        "anli_r2": 0,
+        "anli_r3": 0,
+        "cb": 1,
+        "copa": 0
+    }
+}