bloomz-p3 / evaluation_val /merged.json
Muennighoff's picture
Add files
aa2424b
raw
history blame
No virus
122 kB
{"amazon_reviews_multi_en": {"prompt_body_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5878}, "template_name": "prompt_body_title_to_star"}, "prompt_review_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5492}, "template_name": "prompt_review_to_star"}, "prompt_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.4018}, "template_name": "prompt_title_to_star"}}, "amazon_reviews_multi_es": {"prompt_body_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.53}, "template_name": "prompt_body_title_to_star"}, "prompt_review_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5042}, "template_name": "prompt_review_to_star"}, "prompt_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.37}, "template_name": "prompt_title_to_star"}}, "amazon_reviews_multi_fr": {"prompt_body_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "fr", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5294}, "template_name": "prompt_body_title_to_star"}, "prompt_review_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "fr", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.514}, "template_name": "prompt_review_to_star"}, "prompt_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "fr", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.3664}, "template_name": "prompt_title_to_star"}}, "amazon_reviews_multi_zh": {"prompt_body_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "zh", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.503}, "template_name": "prompt_body_title_to_star"}, "prompt_review_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "zh", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.4882}, "template_name": "prompt_review_to_star"}, "prompt_title_to_star": {"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='amazon_reviews_multi', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "zh", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.3656}, "template_name": "prompt_title_to_star"}}, "aqua_rat_raw": {"Answer questions from options": {"arguments": "Namespace(config_name=None, dataset_config_name='raw', dataset_name='aqua_rat', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Answer questions from options', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "raw", "dataset_name": "aqua_rat", "evaluation": {"accuracy": 0.2283464566929134}, "template_name": "Answer questions from options"}, "answer_quiz": {"arguments": "Namespace(config_name=None, dataset_config_name='raw', dataset_name='aqua_rat', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='answer_quiz', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "raw", "dataset_name": "aqua_rat", "evaluation": {"accuracy": 0.23228346456692914}, "template_name": "answer_quiz"}, "select_the_best_option": {"arguments": "Namespace(config_name=None, dataset_config_name='raw', dataset_name='aqua_rat', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='select_the_best_option', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "raw", "dataset_name": "aqua_rat", "evaluation": {"accuracy": 0.2125984251968504}, "template_name": "select_the_best_option"}}, "art_None": {"choose_hypothesis": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='choose_hypothesis', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "art", "evaluation": {"accuracy": 0.6527415143603134}, "template_name": "choose_hypothesis"}, "choose_hypothesis_believable": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='choose_hypothesis_believable', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "art", "evaluation": {"accuracy": 0.6586161879895561}, "template_name": "choose_hypothesis_believable"}, "choose_hypothesis_desc": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='choose_hypothesis_desc', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "art", "evaluation": {"accuracy": 0.5718015665796344}, "template_name": "choose_hypothesis_desc"}, "choose_hypothesis_likely": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='choose_hypothesis_likely', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "art", "evaluation": {"accuracy": 0.5535248041775457}, "template_name": "choose_hypothesis_likely"}, "choose_hypothesis_options": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='choose_hypothesis_options', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "art", "evaluation": {"accuracy": 0.6298955613577023}, "template_name": "choose_hypothesis_options"}}, "banking77_None": {"direct_to_which_department": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='banking77', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='direct_to_which_department', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "banking77", "evaluation": {"accuracy": 0.22077922077922077}, "template_name": "direct_to_which_department"}, "help_page_topic": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='banking77', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='help_page_topic', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "banking77", "evaluation": {"accuracy": 0.3198051948051948}, "template_name": "help_page_topic"}, "rephrase_as_banking_term": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='banking77', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='rephrase_as_banking_term', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "banking77", "evaluation": {"accuracy": 0.31915584415584414}, "template_name": "rephrase_as_banking_term"}}, "blbooksgenre_title_genre_classifiction": {"classify": {"arguments": "Namespace(config_name=None, dataset_config_name='title_genre_classifiction', dataset_name='blbooksgenre', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='classify', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "title_genre_classifiction", "dataset_name": "blbooksgenre", "evaluation": {"accuracy": 0.25172811059907835}, "template_name": "classify"}, "multi-choice": {"arguments": "Namespace(config_name=None, dataset_config_name='title_genre_classifiction', dataset_name='blbooksgenre', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='multi-choice', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "title_genre_classifiction", "dataset_name": "blbooksgenre", "evaluation": {"accuracy": 0.5161290322580645}, "template_name": "multi-choice"}, "premise_context_first": {"arguments": "Namespace(config_name=None, dataset_config_name='title_genre_classifiction', dataset_name='blbooksgenre', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='premise_context_first', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "title_genre_classifiction", "dataset_name": "blbooksgenre", "evaluation": {"accuracy": 0.7517281105990783}, "template_name": "premise_context_first"}}, "blimp_adjunct_island": {"grammatical_between_1_2": {"arguments": "Namespace(config_name=None, dataset_config_name='adjunct_island', dataset_name='blimp', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='grammatical_between_1_2', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "adjunct_island", "dataset_name": "blimp", "evaluation": {"accuracy": 0.519}, "template_name": "grammatical_between_1_2"}, "grammatical_between_A_B": {"arguments": "Namespace(config_name=None, dataset_config_name='adjunct_island', dataset_name='blimp', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='grammatical_between_A_B', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "adjunct_island", "dataset_name": "blimp", "evaluation": {"accuracy": 0.36}, "template_name": "grammatical_between_A_B"}, "grammatical_which_one_1_2": {"arguments": "Namespace(config_name=None, dataset_config_name='adjunct_island', dataset_name='blimp', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='grammatical_which_one_1_2', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "adjunct_island", "dataset_name": "blimp", "evaluation": {"accuracy": 0.506}, "template_name": "grammatical_which_one_1_2"}, "single_sentence_bad_yes_no": {"arguments": "Namespace(config_name=None, dataset_config_name='adjunct_island', dataset_name='blimp', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='single_sentence_bad_yes_no', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "adjunct_island", "dataset_name": "blimp", "evaluation": {"accuracy": 0.527}, "template_name": "single_sentence_bad_yes_no"}, "single_sentence_good_yes_no": {"arguments": "Namespace(config_name=None, dataset_config_name='adjunct_island', dataset_name='blimp', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='single_sentence_good_yes_no', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "adjunct_island", "dataset_name": "blimp", "evaluation": {"accuracy": 0.49}, "template_name": "single_sentence_good_yes_no"}}, "climate_fever_None": {"claim_and_all_supporting_evidences": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='climate_fever', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='claim_and_all_supporting_evidences', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "climate_fever", "evaluation": {"accuracy": 0.23908794788273616}, "template_name": "claim_and_all_supporting_evidences"}, "fifth_evidence_and_claim_itemization": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='climate_fever', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='fifth_evidence_and_claim_itemization', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "climate_fever", "evaluation": {"accuracy": 0.43713355048859937}, "template_name": "fifth_evidence_and_claim_itemization"}, "first_evidence_and_claim_itemization": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='climate_fever', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='first_evidence_and_claim_itemization', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "climate_fever", "evaluation": {"accuracy": 0.21172638436482086}, "template_name": "first_evidence_and_claim_itemization"}, "second_evidence_and_claim_itemization": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='climate_fever', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='second_evidence_and_claim_itemization', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "climate_fever", "evaluation": {"accuracy": 0.4}, "template_name": "second_evidence_and_claim_itemization"}, "third_evidence_claim_pair": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='climate_fever', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='third_evidence_claim_pair', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "climate_fever", "evaluation": {"accuracy": 0.36351791530944627}, "template_name": "third_evidence_claim_pair"}}, "codah_codah": {"affirmative_instruction_after_sentence_and_choices": {"arguments": "Namespace(config_name=None, dataset_config_name='codah', dataset_name='codah', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='affirmative_instruction_after_sentence_and_choices', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "codah", "dataset_name": "codah", "evaluation": {"accuracy": 0.7917867435158501}, "template_name": "affirmative_instruction_after_sentence_and_choices"}, "affirmative_instruction_before_sentence_and_choices": {"arguments": "Namespace(config_name=None, dataset_config_name='codah', dataset_name='codah', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='affirmative_instruction_before_sentence_and_choices', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "codah", "dataset_name": "codah", "evaluation": {"accuracy": 0.7921469740634006}, "template_name": "affirmative_instruction_before_sentence_and_choices"}, "interrogative_instruction_after_sentence_and_choices": {"arguments": "Namespace(config_name=None, dataset_config_name='codah', dataset_name='codah', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='interrogative_instruction_after_sentence_and_choices', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "codah", "dataset_name": "codah", "evaluation": {"accuracy": 0.797550432276657}, "template_name": "interrogative_instruction_after_sentence_and_choices"}}, "commonsense_qa_None": {"answer_given_question_without_options": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='commonsense_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='answer_given_question_without_options', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "commonsense_qa", "evaluation": {"accuracy": 0.6961506961506961}, "template_name": "answer_given_question_without_options"}, "most_suitable_answer": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='commonsense_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='most_suitable_answer', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "commonsense_qa", "evaluation": {"accuracy": 0.7583947583947583}, "template_name": "most_suitable_answer"}, "question_answering": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='commonsense_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='question_answering', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "commonsense_qa", "evaluation": {"accuracy": 0.7444717444717445}, "template_name": "question_answering"}}, "conv_ai_3_None": {"ambiguous": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='conv_ai_3', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='ambiguous', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "conv_ai_3", "evaluation": {"accuracy": 0.39040207522697795}, "template_name": "ambiguous"}, "clarification_needed": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='conv_ai_3', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='clarification_needed', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "conv_ai_3", "evaluation": {"accuracy": 0.39040207522697795}, "template_name": "clarification_needed"}, "directly_answer": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='conv_ai_3', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='directly_answer', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "conv_ai_3", "evaluation": {"accuracy": 0.6095979247730221}, "template_name": "directly_answer"}, "score_give_number": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='conv_ai_3', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='score_give_number', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "conv_ai_3", "evaluation": {"accuracy": 0.21444012105490703}, "template_name": "score_give_number"}, "score_how_much": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='conv_ai_3', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='score_how_much', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "conv_ai_3", "evaluation": {"accuracy": 0.21444012105490703}, "template_name": "score_how_much"}}, "craigslist_bargains_None": {"best deal": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='craigslist_bargains', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='best deal', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "craigslist_bargains", "evaluation": {"accuracy": 0.5142378559463987}, "template_name": "best deal"}, "good deal for seller": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='craigslist_bargains', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='good deal for seller', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "craigslist_bargains", "evaluation": {"accuracy": 0.2663316582914573}, "template_name": "good deal for seller"}, "good deal for seller no list price": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='craigslist_bargains', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='good deal for seller no list price', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "craigslist_bargains", "evaluation": {"accuracy": 0.135678391959799}, "template_name": "good deal for seller no list price"}, "good deal for seller no list price implicit": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='craigslist_bargains', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='good deal for seller no list price implicit', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "craigslist_bargains", "evaluation": {"accuracy": 0.23785594639865998}, "template_name": "good deal for seller no list price implicit"}}, "emotion_None": {"answer_question_with_emotion_label": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='emotion', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='answer_question_with_emotion_label', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "emotion", "evaluation": {"accuracy": 0.4005}, "template_name": "answer_question_with_emotion_label"}, "answer_with_class_label": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='emotion', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='answer_with_class_label', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "emotion", "evaluation": {"accuracy": 0.3595}, "template_name": "answer_with_class_label"}, "choose_the_best_emotion_label": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='emotion', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='choose_the_best_emotion_label', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "emotion", "evaluation": {"accuracy": 0.4435}, "template_name": "choose_the_best_emotion_label"}, "reply_with_emoation_label": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='emotion', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='reply_with_emoation_label', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "emotion", "evaluation": {"accuracy": 0.469}, "template_name": "reply_with_emoation_label"}}, "financial_phrasebank_sentences_allagree": {"bullish_neutral_bearish": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='bullish_neutral_bearish', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sentences_allagree", "dataset_name": "financial_phrasebank", "evaluation": {"accuracy": 0.3365724381625442}, "template_name": "bullish_neutral_bearish"}, "complementary_industries": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='complementary_industries', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sentences_allagree", "dataset_name": "financial_phrasebank", "evaluation": {"accuracy": 0.09143109540636042}, "template_name": "complementary_industries"}, "sentiment": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='sentiment', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sentences_allagree", "dataset_name": "financial_phrasebank", "evaluation": {"accuracy": 0.36130742049469966}, "template_name": "sentiment"}, "share_price_option": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='share_price_option', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sentences_allagree", "dataset_name": "financial_phrasebank", "evaluation": {"accuracy": 0.36925795053003535}, "template_name": "share_price_option"}, "word_comes_to_mind": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='word_comes_to_mind', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sentences_allagree", "dataset_name": "financial_phrasebank", "evaluation": {"accuracy": 0.03931095406360424}, "template_name": "word_comes_to_mind"}}, "glue_cola": {"Following sentence acceptable": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Following sentence acceptable', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "cola", "dataset_name": "glue", "evaluation": {"accuracy": 0.4506232023010546}, "template_name": "Following sentence acceptable"}, "Make sense yes no": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Make sense yes no', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "cola", "dataset_name": "glue", "evaluation": {"accuracy": 0.3595397890699904}, "template_name": "Make sense yes no"}, "Previous sentence acceptable": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Previous sentence acceptable', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "cola", "dataset_name": "glue", "evaluation": {"accuracy": 0.3096836049856184}, "template_name": "Previous sentence acceptable"}, "editing": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='editing', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "cola", "dataset_name": "glue", "evaluation": {"accuracy": 0.3202301054650048}, "template_name": "editing"}, "is_this_correct": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is_this_correct', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "cola", "dataset_name": "glue", "evaluation": {"accuracy": 0.3288590604026846}, "template_name": "is_this_correct"}}, "glue_sst2": {"following positive negative": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='following positive negative', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sst2", "dataset_name": "glue", "evaluation": {"accuracy": 0.9541284403669725}, "template_name": "following positive negative"}, "happy or mad": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='happy or mad', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sst2", "dataset_name": "glue", "evaluation": {"accuracy": 0.9197247706422018}, "template_name": "happy or mad"}, "positive negative after": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='positive negative after', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sst2", "dataset_name": "glue", "evaluation": {"accuracy": 0.9472477064220184}, "template_name": "positive negative after"}, "review": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='review', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sst2", "dataset_name": "glue", "evaluation": {"accuracy": 0.9541284403669725}, "template_name": "review"}, "said": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='said', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sst2", "dataset_name": "glue", "evaluation": {"accuracy": 0.8555045871559633}, "template_name": "said"}}, "head_qa_en": {"multiple_choice_a_and_q_en": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_a_and_q_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.3374816983894583}, "template_name": "multiple_choice_a_and_q_en"}, "multiple_choice_a_and_q_with_context_en": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_a_and_q_with_context_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.32723279648609077}, "template_name": "multiple_choice_a_and_q_with_context_en"}, "multiple_choice_q_and_a_en": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_q_and_a_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.47144948755490484}, "template_name": "multiple_choice_q_and_a_en"}, "multiple_choice_q_and_a_index_en": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_q_and_a_index_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.4538799414348463}, "template_name": "multiple_choice_q_and_a_index_en"}, "multiple_choice_q_and_a_index_with_context_en": {"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_q_and_a_index_with_context_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.45095168374816985}, "template_name": "multiple_choice_q_and_a_index_with_context_en"}}, "head_qa_es": {"multiple_choice_a_and_q_en": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_a_and_q_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.32503660322108346}, "template_name": "multiple_choice_a_and_q_en"}, "multiple_choice_a_and_q_with_context_en": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_a_and_q_with_context_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.32210834553440704}, "template_name": "multiple_choice_a_and_q_with_context_en"}, "multiple_choice_q_and_a_en": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_q_and_a_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.4685212298682284}, "template_name": "multiple_choice_q_and_a_en"}, "multiple_choice_q_and_a_index_en": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_q_and_a_index_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.4289897510980966}, "template_name": "multiple_choice_q_and_a_index_en"}, "multiple_choice_q_and_a_index_with_context_en": {"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='head_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='multiple_choice_q_and_a_index_with_context_en', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "head_qa", "evaluation": {"accuracy": 0.4333821376281113}, "template_name": "multiple_choice_q_and_a_index_with_context_en"}}, "health_fact_None": {"claim_explanation_classification": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='health_fact', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='claim_explanation_classification', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "health_fact", "evaluation": {"accuracy": 0.49551020408163265}, "template_name": "claim_explanation_classification"}, "claim_veracity_classification_after_reading_I_believe": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='health_fact', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='claim_veracity_classification_after_reading_I_believe', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "health_fact", "evaluation": {"accuracy": 0.21551020408163266}, "template_name": "claim_veracity_classification_after_reading_I_believe"}, "claim_veracity_classification_tell_me": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='health_fact', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=1, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='claim_veracity_classification_tell_me', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "health_fact", "evaluation": {"accuracy": 0.05142857142857143}, "template_name": "claim_veracity_classification_tell_me"}}, "hlgd_None": {"is_same_event_editor_asks": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='hlgd', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is_same_event_editor_asks', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "hlgd", "evaluation": {"accuracy": 0.6872885451909135}, "template_name": "is_same_event_editor_asks"}, "is_same_event_interrogative_talk": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='hlgd', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is_same_event_interrogative_talk', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "hlgd", "evaluation": {"accuracy": 0.6795553407443209}, "template_name": "is_same_event_interrogative_talk"}, "is_same_event_refer": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='hlgd', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is_same_event_refer', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "hlgd", "evaluation": {"accuracy": 0.8221362977283712}, "template_name": "is_same_event_refer"}, "is_same_event_with_time_interrogative_related": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='hlgd', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is_same_event_with_time_interrogative_related', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "hlgd", "evaluation": {"accuracy": 0.8144030932817786}, "template_name": "is_same_event_with_time_interrogative_related"}, "is_same_event_with_time_interrogative_talk": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='hlgd', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is_same_event_with_time_interrogative_talk', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "hlgd", "evaluation": {"accuracy": 0.8090865152247463}, "template_name": "is_same_event_with_time_interrogative_talk"}}, "hyperpartisan_news_detection_byarticle": {"consider_does_it_follow_a_hyperpartisan_argumentation": {"arguments": "Namespace(config_name=None, dataset_config_name='byarticle', dataset_name='hyperpartisan_news_detection', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='consider_does_it_follow_a_hyperpartisan_argumentation', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "byarticle", "dataset_name": "hyperpartisan_news_detection", "evaluation": {"accuracy": 0.6170542635658914}, "template_name": "consider_does_it_follow_a_hyperpartisan_argumentation"}, "consider_it_exhibits_extreme_one_sidedness": {"arguments": "Namespace(config_name=None, dataset_config_name='byarticle', dataset_name='hyperpartisan_news_detection', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='consider_it_exhibits_extreme_one_sidedness', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "byarticle", "dataset_name": "hyperpartisan_news_detection", "evaluation": {"accuracy": 0.6310077519379845}, "template_name": "consider_it_exhibits_extreme_one_sidedness"}, "consume_with_caution": {"arguments": "Namespace(config_name=None, dataset_config_name='byarticle', dataset_name='hyperpartisan_news_detection', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='consume_with_caution', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "byarticle", "dataset_name": "hyperpartisan_news_detection", "evaluation": {"accuracy": 0.6248062015503876}, "template_name": "consume_with_caution"}, "extreme_left_wing_or_right_wing": {"arguments": "Namespace(config_name=None, dataset_config_name='byarticle', dataset_name='hyperpartisan_news_detection', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='extreme_left_wing_or_right_wing', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "byarticle", "dataset_name": "hyperpartisan_news_detection", "evaluation": {"accuracy": 0.6186046511627907}, "template_name": "extreme_left_wing_or_right_wing"}, "follows_hyperpartisan_argumentation": {"arguments": "Namespace(config_name=None, dataset_config_name='byarticle', dataset_name='hyperpartisan_news_detection', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='follows_hyperpartisan_argumentation', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "byarticle", "dataset_name": "hyperpartisan_news_detection", "evaluation": {"accuracy": 0.627906976744186}, "template_name": "follows_hyperpartisan_argumentation"}}, "liar_None": {"Given statement guess category": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='liar', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Given statement guess category', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "liar", "evaluation": {"accuracy": 0.20249221183800623}, "template_name": "Given statement guess category"}}, "lince_sa_spaeng": {"express sentiment": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='express sentiment', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sa_spaeng", "dataset_name": "lince", "evaluation": {"accuracy": 0.5874125874125874}, "template_name": "express sentiment"}, "negation template": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='negation template', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sa_spaeng", "dataset_name": "lince", "evaluation": {"accuracy": 0.5879505110274341}, "template_name": "negation template"}, "original poster expressed sentiment": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='original poster expressed sentiment', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sa_spaeng", "dataset_name": "lince", "evaluation": {"accuracy": 0.5901022054868209}, "template_name": "original poster expressed sentiment"}, "sentiment trying to express": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='sentiment trying to express', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sa_spaeng", "dataset_name": "lince", "evaluation": {"accuracy": 0.5809575040344271}, "template_name": "sentiment trying to express"}, "the author seem": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='the author seem', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "sa_spaeng", "dataset_name": "lince", "evaluation": {"accuracy": 0.5911780527165142}, "template_name": "the author seem"}}, "math_qa_None": {"choose_correct_og": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='choose_correct_og', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "math_qa", "evaluation": {"accuracy": 0.24489112227805696}, "template_name": "choose_correct_og"}, "first_choice_then_problem": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='first_choice_then_problem', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "math_qa", "evaluation": {"accuracy": 0.2020100502512563}, "template_name": "first_choice_then_problem"}, "gre_problem": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='gre_problem', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "math_qa", "evaluation": {"accuracy": 0.22981574539363483}, "template_name": "gre_problem"}, "pick_the_correct": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='pick_the_correct', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "math_qa", "evaluation": {"accuracy": 0.23819095477386934}, "template_name": "pick_the_correct"}, "problem_set_type": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='test', target_max_length=256, template_config_name=None, template_name='problem_set_type', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "math_qa", "evaluation": {"accuracy": 0.4824120603015075}, "template_name": "problem_set_type"}}, "mlsum_es": {"layman_summ_es": {"bleu": 2.7240547697036885, "bleu_stderr": 0.09873871923462486, "rouge1_fmeasure": 0.2152514623131688, "rouge1_fmeasure_stderr": 0.00235779670443836, "rouge1_precision": 0.20922341031018796, "rouge1_precision_stderr": 0.0025375595568407133, "rouge1_recall": 0.2686473017701524, "rouge1_recall_stderr": 0.003243235357501857, "rouge2_fmeasure": 0.05799389954948545, "rouge2_fmeasure_stderr": 0.0016274426630508128, "rouge2_precision": 0.05570540199745807, "rouge2_precision_stderr": 0.0017110691355917763, "rouge2_recall": 0.07438656912468752, "rouge2_recall_stderr": 0.0021166092536387365, "rougeL_fmeasure": 0.16039478846940786, "rougeL_fmeasure_stderr": 0.001896226389379961, "rougeL_precision": 0.15759931030583763, "rougeL_precision_stderr": 0.002106134774825538, "rougeL_recall": 0.19930722307414317, "rougeL_recall_stderr": 0.002532772355449645, "rougeLsum_fmeasure": 0.16743568643555828, "rougeLsum_fmeasure_stderr": 0.0019537523119573046, "rougeLsum_precision": 0.16395293631004723, "rougeLsum_precision_stderr": 0.002154082917851902, "rougeLsum_recall": 0.20952028063185008, "rougeLsum_recall_stderr": 0.0026942495945950105}, "palm_prompt": {"bleu": 3.916653973278921, "bleu_stderr": 0.13320174843239593, "rouge1_fmeasure": 0.23777639529079925, "rouge1_fmeasure_stderr": 0.002353644348181871, "rouge1_precision": 0.238006133260283, "rouge1_precision_stderr": 0.002592201534618347, "rouge1_recall": 0.28440217742758356, "rouge1_recall_stderr": 0.0032686854143649984, "rouge2_fmeasure": 0.07113316160241126, "rouge2_fmeasure_stderr": 0.0018688817237639986, "rouge2_precision": 0.07002703238460953, "rouge2_precision_stderr": 0.0019467819130961877, "rouge2_recall": 0.08785375976968864, "rouge2_recall_stderr": 0.002327977791822037, "rougeL_fmeasure": 0.17988536856242504, "rougeL_fmeasure_stderr": 0.0020214064930605237, "rougeL_precision": 0.18083602836591542, "rougeL_precision_stderr": 0.002222442145519105, "rougeL_recall": 0.21460635266053055, "rougeL_recall_stderr": 0.0026818176382247703, "rougeLsum_fmeasure": 0.18777680808178215, "rougeLsum_fmeasure_stderr": 0.0020733063902455624, "rougeLsum_precision": 0.18813961119405384, "rougeLsum_precision_stderr": 0.00226738528856828, "rougeLsum_recall": 0.2256886585459008, "rougeLsum_recall_stderr": 0.0028543730103714284}, "summarise_this_in_es_few_sentences": {"bleu": 3.1798181363636027, "bleu_stderr": 0.11186670060245504, "rouge1_fmeasure": 0.22609931277536174, "rouge1_fmeasure_stderr": 0.0022513761751812465, "rouge1_precision": 0.22243683019483318, "rouge1_precision_stderr": 0.0025163867913330275, "rouge1_recall": 0.2852722290772078, "rouge1_recall_stderr": 0.003171172872929976, "rouge2_fmeasure": 0.06124625613676503, "rouge2_fmeasure_stderr": 0.0017334783966114224, "rouge2_precision": 0.05973518770656727, "rouge2_precision_stderr": 0.001893687676123278, "rouge2_recall": 0.07895712586687036, "rouge2_recall_stderr": 0.0021825236706687757, "rougeL_fmeasure": 0.16955218451285564, "rougeL_fmeasure_stderr": 0.0018846082416891987, "rougeL_precision": 0.16887516659649215, "rougeL_precision_stderr": 0.002178357265241399, "rougeL_recall": 0.21318196705739953, "rougeL_recall_stderr": 0.002555636899506184, "rougeLsum_fmeasure": 0.17665025797213654, "rougeLsum_fmeasure_stderr": 0.0019283819849723594, "rougeLsum_precision": 0.1753893170918232, "rougeLsum_precision_stderr": 0.002217858281060236, "rougeLsum_recall": 0.22340887603027765, "rougeLsum_recall_stderr": 0.0027033960043687556}}, "movie_rationales_None": {"Evidences + review": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='movie_rationales', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Evidences + review', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "movie_rationales", "evaluation": {"accuracy": 0.98}, "template_name": "Evidences + review"}, "Evidences sentiment classification": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='movie_rationales', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Evidences sentiment classification', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "movie_rationales", "evaluation": {"accuracy": 0.995}, "template_name": "Evidences sentiment classification"}, "Standard binary sentiment analysis": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='movie_rationales', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Standard binary sentiment analysis', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "movie_rationales", "evaluation": {"accuracy": 0.97}, "template_name": "Standard binary sentiment analysis"}}, "mwsc_None": {"in-the-sentence": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='in-the-sentence', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "mwsc", "evaluation": {"accuracy": 0.7804878048780488}, "template_name": "in-the-sentence"}, "in-the-sentence-question-first": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='in-the-sentence-question-first', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "mwsc", "evaluation": {"accuracy": 0.7195121951219512}, "template_name": "in-the-sentence-question-first"}, "is-correct": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is-correct', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "mwsc", "evaluation": {"accuracy": 0.524390243902439}, "template_name": "is-correct"}, "options-or": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='options-or', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "mwsc", "evaluation": {"accuracy": 0.7926829268292683}, "template_name": "options-or"}, "what-think": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='what-think', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "mwsc", "evaluation": {"accuracy": 0.8048780487804879}, "template_name": "what-think"}}, "onestop_english_None": {"ara_context": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='ara_context', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "onestop_english", "evaluation": {"accuracy": 0.6490299823633157}, "template_name": "ara_context"}, "assess": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='assess', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "onestop_english", "evaluation": {"accuracy": 0.5802469135802469}, "template_name": "assess"}, "determine_reading_level_from_the_first_three_sentences": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='determine_reading_level_from_the_first_three_sentences', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "onestop_english", "evaluation": {"accuracy": 0.6596119929453262}, "template_name": "determine_reading_level_from_the_first_three_sentences"}, "esl_context": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='esl_context', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "onestop_english", "evaluation": {"accuracy": 0.562610229276896}, "template_name": "esl_context"}, "esl_variation": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='esl_variation', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "onestop_english", "evaluation": {"accuracy": 0.6261022927689595}, "template_name": "esl_variation"}}, "poem_sentiment_None": {"guess_sentiment_without_options_variation_1": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='poem_sentiment', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='guess_sentiment_without_options_variation_1', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "poem_sentiment", "evaluation": {"accuracy": 0.24761904761904763}, "template_name": "guess_sentiment_without_options_variation_1"}, "most_appropriate_sentiment": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='poem_sentiment', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='most_appropriate_sentiment', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "poem_sentiment", "evaluation": {"accuracy": 0.26666666666666666}, "template_name": "most_appropriate_sentiment"}, "positive_or_negative_sentiment_variation_1": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='poem_sentiment', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='positive_or_negative_sentiment_variation_1', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "poem_sentiment", "evaluation": {"accuracy": 0.2571428571428571}, "template_name": "positive_or_negative_sentiment_variation_1"}, "positive_or_negative_sentiment_variation_2": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='poem_sentiment', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='positive_or_negative_sentiment_variation_2', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "poem_sentiment", "evaluation": {"accuracy": 0.22857142857142856}, "template_name": "positive_or_negative_sentiment_variation_2"}, "question_answer_format": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='poem_sentiment', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='question_answer_format', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "poem_sentiment", "evaluation": {"accuracy": 0.26666666666666666}, "template_name": "question_answer_format"}}, "pubmed_qa_pqa_labeled": {"Long Answer to Final Decision": {"arguments": "Namespace(config_name=None, dataset_config_name='pqa_labeled', dataset_name='pubmed_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='Long Answer to Final Decision', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "pqa_labeled", "dataset_name": "pubmed_qa", "evaluation": {"accuracy": 0.698}, "template_name": "Long Answer to Final Decision"}, "Question Answering (Short)": {"arguments": "Namespace(config_name=None, dataset_config_name='pqa_labeled', dataset_name='pubmed_qa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='Question Answering (Short)', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "pqa_labeled", "dataset_name": "pubmed_qa", "evaluation": {"accuracy": 0.741}, "template_name": "Question Answering (Short)"}}, "riddle_sense_None": {"answer_given_question_without_options": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='riddle_sense', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='answer_given_question_without_options', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "riddle_sense", "evaluation": {"accuracy": 0.5337904015670911}, "template_name": "answer_given_question_without_options"}, "most_suitable_answer": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='riddle_sense', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='most_suitable_answer', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "riddle_sense", "evaluation": {"accuracy": 0.5288932419196866}, "template_name": "most_suitable_answer"}, "question_answering": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='riddle_sense', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='question_answering', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "riddle_sense", "evaluation": {"accuracy": 0.4975514201762977}, "template_name": "question_answering"}, "question_to_answer_index": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='riddle_sense', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='question_to_answer_index', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "riddle_sense", "evaluation": {"accuracy": 0.5239960822722821}, "template_name": "question_to_answer_index"}}, "scicite_None": {"Classify intent": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='scicite', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Classify intent', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "scicite", "evaluation": {"accuracy": 0.3078602620087336}, "template_name": "Classify intent"}, "Classify intent (choices first)": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='scicite', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Classify intent (choices first)', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "scicite", "evaluation": {"accuracy": 0.16921397379912664}, "template_name": "Classify intent (choices first)"}, "Classify intent (select choice)": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='scicite', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Classify intent (select choice)', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "scicite", "evaluation": {"accuracy": 0.3067685589519651}, "template_name": "Classify intent (select choice)"}, "Classify intent w/section (select choice)": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='scicite', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='Classify intent w/section (select choice)', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "scicite", "evaluation": {"accuracy": 0.38427947598253276}, "template_name": "Classify intent w/section (select choice)"}, "can_describe": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='scicite', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='can_describe', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "scicite", "evaluation": {"accuracy": 0.34497816593886466}, "template_name": "can_describe"}}, "selqa_answer_selection_analysis": {"is-he-talking-about": {"arguments": "Namespace(config_name=None, dataset_config_name='answer_selection_analysis', dataset_name='selqa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='is-he-talking-about', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "answer_selection_analysis", "dataset_name": "selqa", "evaluation": {"accuracy": 0.913375796178344}, "template_name": "is-he-talking-about"}, "make-sense-rand": {"arguments": "Namespace(config_name=None, dataset_config_name='answer_selection_analysis', dataset_name='selqa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='make-sense-rand', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "answer_selection_analysis", "dataset_name": "selqa", "evaluation": {"accuracy": 0.9452229299363057}, "template_name": "make-sense-rand"}, "which-answer-1st-vs-random": {"arguments": "Namespace(config_name=None, dataset_config_name='answer_selection_analysis', dataset_name='selqa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='which-answer-1st-vs-random', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "answer_selection_analysis", "dataset_name": "selqa", "evaluation": {"accuracy": 0.8267515923566879}, "template_name": "which-answer-1st-vs-random"}, "would-make-sense-qu-rand": {"arguments": "Namespace(config_name=None, dataset_config_name='answer_selection_analysis', dataset_name='selqa', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='would-make-sense-qu-rand', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "answer_selection_analysis", "dataset_name": "selqa", "evaluation": {"accuracy": 0.9095541401273886}, "template_name": "would-make-sense-qu-rand"}}, "snips_built_in_intents_None": {"categorize_query": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='categorize_query', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "snips_built_in_intents", "evaluation": {"accuracy": 0.7926829268292683}, "template_name": "categorize_query"}, "categorize_query_brief": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='categorize_query_brief', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "snips_built_in_intents", "evaluation": {"accuracy": 0.7195121951219512}, "template_name": "categorize_query_brief"}, "intent_query": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='intent_query', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "snips_built_in_intents", "evaluation": {"accuracy": 0.4268292682926829}, "template_name": "intent_query"}, "query_intent": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='query_intent', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "snips_built_in_intents", "evaluation": {"accuracy": 0.7408536585365854}, "template_name": "query_intent"}, "voice_intent": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/p31lossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=4, prefixlm=False, split='train', target_max_length=256, template_config_name=None, template_name='voice_intent', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "snips_built_in_intents", "evaluation": {"accuracy": 0.7652439024390244}, "template_name": "voice_intent"}}, "wmt14_fr_en_en-fr": {"a_good_translation-en-fr-source+target": {"bleu": 5.708641852963801, "bleu_stderr": 0.08616285971552679}, "a_good_translation-en-fr-target": {"bleu": 2.1803001724318145, "bleu_stderr": 0.11385262619041567}, "gpt3-en-fr": {"bleu": 13.227227858954294, "bleu_stderr": 0.4152455632280683}, "version-en-fr-target": {"bleu": 3.76456058717039, "bleu_stderr": 0.15671670767309734}, "xglm-en-fr-target": {"bleu": 1.3771806987654605, "bleu_stderr": 0.09711955435659027}}, "wmt14_fr_en_fr-en": {"a_good_translation-fr-en-source+target": {"bleu": 35.68647218780612, "bleu_stderr": 0.39526448750149445}, "a_good_translation-fr-en-target": {"bleu": 35.195872833563755, "bleu_stderr": 0.33749050265136066}, "gpt3-fr-en": {"bleu": 32.55636093987592, "bleu_stderr": 0.3317275309610832}, "version-fr-en-target": {"bleu": 33.91775518002708, "bleu_stderr": 0.3049303286108613}, "xglm-fr-en-target": {"bleu": 12.924351243878487, "bleu_stderr": 0.1518176622463832}}, "wmt14_hi_en_en-hi": {"a_good_translation-en-hi-source+target": {"bleu": 0.6821708583642451, "bleu_stderr": 0.0763974720474171}, "a_good_translation-en-hi-target": {"bleu": 0.20182870719270515, "bleu_stderr": 0.035326482832351895}, "gpt-3-en-hi-target": {"bleu": 0.48381042997152507, "bleu_stderr": 0.05651687919190645}, "version-en-hi-target": {"bleu": 0.45496025805613965, "bleu_stderr": 0.10794468735965664}, "xglm-en-hi-target": {"bleu": 0.6476009503302996, "bleu_stderr": 0.05837959092449826}}, "wmt14_hi_en_hi-en": {"a_good_translation-hi-en-source+target": {"bleu": 24.59406560241934, "bleu_stderr": 0.2882480050419643}, "a_good_translation-hi-en-target": {"bleu": 24.756498535867514, "bleu_stderr": 0.3043833206604718}, "gpt-3-hi-en-target": {"bleu": 20.14199423501091, "bleu_stderr": 0.4057111103064145}, "version-hi-en-target": {"bleu": 22.687582168757377, "bleu_stderr": 0.27745199363271594}, "xglm-hi-en-target": {"bleu": 4.035937869116407, "bleu_stderr": 0.2515310144266859}}}