bloomz-7b1-p3/evaluation_bloomz-7b1-p3/evaluation_val/merged.json

{"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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.6176}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5592}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "en", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.3922}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5526}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5296}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "es", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.3646}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "fr", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5332}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "fr", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5182}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "fr", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.3644}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_body_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "zh", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5174}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_review_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "zh", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.5006}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='prompt_title_to_star', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": "zh", "dataset_name": "amazon_reviews_multi", "evaluation": {"accuracy": 0.3874}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.24015748031496062}, "template_name": "Answer questions from options"}, "answer_quiz": {"arguments": "Namespace(config_name=None, dataset_config_name='raw', dataset_name='aqua_rat', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.22440944881889763}, "template_name": "answer_quiz"}, "select_the_best_option": {"arguments": "Namespace(config_name=None, dataset_config_name='raw', dataset_name='aqua_rat', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2559055118110236}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5926892950391645}, "template_name": "choose_hypothesis"}, "choose_hypothesis_believable": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5711488250652742}, "template_name": "choose_hypothesis_believable"}, "choose_hypothesis_desc": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5169712793733682}, "template_name": "choose_hypothesis_desc"}, "choose_hypothesis_likely": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5300261096605744}, "template_name": "choose_hypothesis_likely"}, "choose_hypothesis_options": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='art', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5672323759791122}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='direct_to_which_department', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "banking77", "evaluation": {"accuracy": 0.16753246753246753}, "template_name": "direct_to_which_department"}, "help_page_topic": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='banking77', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='help_page_topic', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "banking77", "evaluation": {"accuracy": 0.26785714285714285}, "template_name": "help_page_topic"}, "rephrase_as_banking_term": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='banking77', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='rephrase_as_banking_term', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "banking77", "evaluation": {"accuracy": 0.274025974025974}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.25057603686635943}, "template_name": "classify"}, "multi-choice": {"arguments": "Namespace(config_name=None, dataset_config_name='title_genre_classifiction', dataset_name='blbooksgenre', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.25057603686635943}, "template_name": "multi-choice"}, "premise_context_first": {"arguments": "Namespace(config_name=None, dataset_config_name='title_genre_classifiction', dataset_name='blbooksgenre', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7321428571428571}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.512}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.464}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.512}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.52}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.493}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='claim_and_all_supporting_evidences', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "climate_fever", "evaluation": {"accuracy": 0.3166123778501629}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.4749185667752443}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.22996742671009773}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.24625407166123778}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.24234527687296417}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6693083573487032}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6509365994236311}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6761527377521613}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6388206388206388}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7313677313677314}, "template_name": "most_suitable_answer"}, "question_answering": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='commonsense_qa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7158067158067158}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.057933419801124084}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.010376134889753566}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='best deal', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "craigslist_bargains", "evaluation": {"accuracy": 0.5192629815745393}, "template_name": "best deal"}, "good deal for seller": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='craigslist_bargains', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2529313232830821}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.09715242881072027}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.24623115577889448}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3375}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.214}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.312}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.4495}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3878091872791519}, "template_name": "bullish_neutral_bearish"}, "complementary_industries": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.10114840989399293}, "template_name": "complementary_industries"}, "sentiment": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.35644876325088337}, "template_name": "sentiment"}, "share_price_option": {"arguments": "Namespace(config_name=None, dataset_config_name='sentences_allagree', dataset_name='financial_phrasebank', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3670494699646643}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.08259717314487633}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.37583892617449666}, "template_name": "Following sentence acceptable"}, "Make sense yes no": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.33940556088207097}, "template_name": "Make sense yes no"}, "Previous sentence acceptable": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.31255992329817833}, "template_name": "Previous sentence acceptable"}, "editing": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3844678811121764}, "template_name": "editing"}, "is_this_correct": {"arguments": "Namespace(config_name=None, dataset_config_name='cola', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.37775647171620325}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.9426605504587156}, "template_name": "following positive negative"}, "happy or mad": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.8279816513761468}, "template_name": "happy or mad"}, "positive negative after": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.9254587155963303}, "template_name": "review"}, "said": {"arguments": "Namespace(config_name=None, dataset_config_name='sst2', dataset_name='glue', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.9059633027522935}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.29428989751098095}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.29502196193265007}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3938506588579795}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.30307467057101023}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.30234260614934116}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2730600292825769}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.27232796486090777}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.36530014641288433}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3074670571010249}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3089311859443631}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5591836734693878}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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_veracity_classification_after_reading_I_believe', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "health_fact", "evaluation": {"accuracy": 0.34938775510204084}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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_veracity_classification_tell_me', tokenizer_name=None, use_slow_tokenizer=False)", "dataset_config_name": null, "dataset_name": "health_fact", "evaluation": {"accuracy": 0.48244897959183675}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6926051232479459}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6582890285161914}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7858869018849686}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7839536007733204}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7786370227162881}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6232558139534884}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6294573643410852}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6077519379844961}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.19314641744548286}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5696611081226466}, "template_name": "express sentiment"}, "negation template": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3851533082302313}, "template_name": "negation template"}, "original poster expressed sentiment": {"arguments": "Namespace(config_name=None, dataset_config_name='sa_spaeng', dataset_name='lince', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5841850457235073}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5771920387305003}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.23484087102177553}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2254606365159129}, "template_name": "first_choice_then_problem"}, "gre_problem": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.21943048576214405}, "template_name": "gre_problem"}, "pick_the_correct": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2338358458961474}, "template_name": "pick_the_correct"}, "problem_set_type": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='math_qa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.29246231155778896}, "template_name": "problem_set_type"}}, "mlsum_es": {"layman_summ_es": {"bleu": 2.6830705121606706, "bleu_stderr": 0.14257713719805254, "rouge1_fmeasure": 0.19665942356583802, "rouge1_fmeasure_stderr": 0.00245186970283176, "rouge1_precision": 0.19994210865731296, "rouge1_precision_stderr": 0.0026547454621461738, "rouge1_recall": 0.22882499765155356, "rouge1_recall_stderr": 0.0030837265632016487, "rouge2_fmeasure": 0.0515077211753521, "rouge2_fmeasure_stderr": 0.0015699354248315028, "rouge2_precision": 0.05189155779128239, "rouge2_precision_stderr": 0.0016867882237885771, "rouge2_recall": 0.06074745104675877, "rouge2_recall_stderr": 0.0018807564961523813, "rougeL_fmeasure": 0.14897381009906005, "rougeL_fmeasure_stderr": 0.0019744371501868186, "rougeL_precision": 0.15286714092332523, "rougeL_precision_stderr": 0.0021813478665272707, "rougeL_recall": 0.17277746632777954, "rougeL_recall_stderr": 0.002444939812221139, "rougeLsum_fmeasure": 0.15454418245332874, "rougeLsum_fmeasure_stderr": 0.0020344070829227297, "rougeLsum_precision": 0.15829244217518917, "rougeLsum_precision_stderr": 0.002241835408305656, "rougeLsum_recall": 0.17998021570910885, "rougeLsum_recall_stderr": 0.002565278973604084}, "palm_prompt": {"bleu": 3.341310161344892, "bleu_stderr": 0.12383760876849086, "rouge1_fmeasure": 0.23872886986952627, "rouge1_fmeasure_stderr": 0.0021217578248352883, "rouge1_precision": 0.2238936517609025, "rouge1_precision_stderr": 0.002299862104308459, "rouge1_recall": 0.31290976115097796, "rouge1_recall_stderr": 0.003157461966656448, "rouge2_fmeasure": 0.06770985280514573, "rouge2_fmeasure_stderr": 0.0016286057939871985, "rouge2_precision": 0.06209477646349353, "rouge2_precision_stderr": 0.001551591733457607, "rouge2_recall": 0.09261450488619867, "rouge2_recall_stderr": 0.002338770523626696, "rougeL_fmeasure": 0.17833059997868725, "rougeL_fmeasure_stderr": 0.0017648367718678965, "rougeL_precision": 0.16744902972078152, "rougeL_precision_stderr": 0.001881884121187265, "rougeL_recall": 0.23426964040901505, "rougeL_recall_stderr": 0.0025983311247854634, "rougeLsum_fmeasure": 0.1876134227034203, "rougeLsum_fmeasure_stderr": 0.0018064283175187946, "rougeLsum_precision": 0.1755707446810662, "rougeLsum_precision_stderr": 0.0019188012583382194, "rougeLsum_recall": 0.24827775226125046, "rougeLsum_recall_stderr": 0.0027727938693488186}, "summarise_this_in_es_few_sentences": {"bleu": 2.2245794650879462, "bleu_stderr": 0.07743700029169612, "rouge1_fmeasure": 0.21126423815884174, "rouge1_fmeasure_stderr": 0.0021771710222460634, "rouge1_precision": 0.18419910608261986, "rouge1_precision_stderr": 0.002301564923577535, "rouge1_recall": 0.33528109600140793, "rouge1_recall_stderr": 0.0034321373331462294, "rouge2_fmeasure": 0.05913033007358818, "rouge2_fmeasure_stderr": 0.0014890254374386052, "rouge2_precision": 0.051112897675373886, "rouge2_precision_stderr": 0.0014393881241720322, "rouge2_recall": 0.09793074579590116, "rouge2_recall_stderr": 0.0024100392963833633, "rougeL_fmeasure": 0.15711042852214044, "rougeL_fmeasure_stderr": 0.0017251643310554304, "rougeL_precision": 0.13714024915254835, "rougeL_precision_stderr": 0.0018260510947169805, "rougeL_recall": 0.2521886801730905, "rougeL_recall_stderr": 0.0027817174072391373, "rougeLsum_fmeasure": 0.1670307582655529, "rougeLsum_fmeasure_stderr": 0.0018288687884684008, "rougeLsum_precision": 0.14518990658432604, "rougeLsum_precision_stderr": 0.001900911801455617, "rougeLsum_recall": 0.26906405645015485, "rougeLsum_recall_stderr": 0.0030454611701701234}}, "movie_rationales_None": {"Evidences + review": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='movie_rationales', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.97}, "template_name": "Evidences + review"}, "Evidences sentiment classification": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='movie_rationales', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 1.0}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.95}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6219512195121951}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5853658536585366}, "template_name": "in-the-sentence-question-first"}, "is-correct": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5365853658536586}, "template_name": "is-correct"}, "options-or": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6097560975609756}, "template_name": "options-or"}, "what-think": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='mwsc', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6097560975609756}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3333333333333333}, "template_name": "ara_context"}, "assess": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3333333333333333}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.5696649029982364}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3333333333333333}, "template_name": "esl_context"}, "esl_variation": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='onestop_english', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3333333333333333}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.22857142857142856}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2571428571428571}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.21904761904761905}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.24761904761904763}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.598}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.581}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.4534769833496572}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.4348677766895201}, "template_name": "most_suitable_answer"}, "question_answering": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='riddle_sense', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.4407443682664055}, "template_name": "question_answering"}, "question_to_answer_index": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='riddle_sense', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3878550440744368}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.15065502183406113}, "template_name": "Classify intent"}, "Classify intent (choices first)": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='scicite', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.1331877729257642}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.2652838427947598}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.3537117903930131}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.15283842794759825}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.9121019108280255}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.9171974522292994}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7503184713375797}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.8993630573248408}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.47865853658536583}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.375}, "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='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.31402439024390244}, "template_name": "intent_query"}, "query_intent": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.7012195121951219}, "template_name": "query_intent"}, "voice_intent": {"arguments": "Namespace(config_name=None, dataset_config_name=None, dataset_name='snips_built_in_intents', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/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": 0.6128048780487805}, "template_name": "voice_intent"}}, "wmt14_fr_en_en-fr": {"a_good_translation-en-fr-source+target": {"bleu": 2.125573406419127, "bleu_stderr": 0.09981676122698169}, "a_good_translation-en-fr-target": {"bleu": 1.5697853682886957, "bleu_stderr": 0.10176333685236229}, "gpt3-en-fr": {"bleu": 0.37928468482204986, "bleu_stderr": 0.03833854862936989}, "version-en-fr-target": {"bleu": 4.788559958687529, "bleu_stderr": 0.12647149552786194}, "xglm-en-fr-target": {"bleu": 2.186171298454336, "bleu_stderr": 0.09641163271059554}}, "wmt14_fr_en_fr-en": {"a_good_translation-fr-en-source+target": {"bleu": 30.388346190168132, "bleu_stderr": 0.28706919566129924}, "a_good_translation-fr-en-target": {"bleu": 22.361703612398195, "bleu_stderr": 0.43872418791072576}, "gpt3-fr-en": {"bleu": 17.167001660570335, "bleu_stderr": 0.3999014258297822}, "version-fr-en-target": {"bleu": 23.925613843737143, "bleu_stderr": 0.2692548707999714}, "xglm-fr-en-target": {"bleu": 14.10190003658709, "bleu_stderr": 0.1974741324240151}}, "wmt14_hi_en_en-hi": {"a_good_translation-en-hi-source+target": {"bleu": 0.18051438917625368, "bleu_stderr": 0.03338441915097909}, "a_good_translation-en-hi-target": {"bleu": 0.1812629246502659, "bleu_stderr": 0.04198901460363051}, "gpt-3-en-hi-target": {"bleu": 0.010782650615890082, "bleu_stderr": 0.003615918191553956}, "version-en-hi-target": {"bleu": 0.1858574511075315, "bleu_stderr": 0.029122685049572238}, "xglm-en-hi-target": {"bleu": 0.002225608801197892, "bleu_stderr": 0.0005988947090265846}}, "wmt14_hi_en_hi-en": {"a_good_translation-hi-en-source+target": {"bleu": 16.056644593701627, "bleu_stderr": 0.2809620281933667}, "a_good_translation-hi-en-target": {"bleu": 15.032491079468809, "bleu_stderr": 0.2628594862835867}, "gpt-3-hi-en-target": {"bleu": 5.607403962346587, "bleu_stderr": 0.26092845447942553}, "version-hi-en-target": {"bleu": 15.167071858881462, "bleu_stderr": 0.2573529636593602}, "xglm-hi-en-target": {"bleu": 3.675518735361532, "bleu_stderr": 0.17101231729659816}}}