Commit afbf8e66 authored by Julen Etxaniz's avatar Julen Etxaniz
Browse files

add more mpt and llama results

parent 0e849758
...@@ -432,6 +432,15 @@ ...@@ -432,6 +432,15 @@
| | |f1 |11.35|± | 0.23| | | |f1 |11.35|± | 0.23|
|race| 1|acc |39.90|± | 1.52| |race| 1|acc |39.90|± | 1.52|
## llama-7B_unscramble_0-shot.json
| Task |Version|Metric|Value| |Stderr|
|----------------|------:|------|----:|---|-----:|
|anagrams1 | 0|acc | 0|± | 0|
|anagrams2 | 0|acc | 0|± | 0|
|cycle_letters | 0|acc | 0|± | 0|
|random_insertion| 0|acc | 0|± | 0|
|reversed_words | 0|acc | 0|± | 0|
## llama-7B_xcopa_0-shot.json ## llama-7B_xcopa_0-shot.json
| Task |Version|Metric|Value| |Stderr| | Task |Version|Metric|Value| |Stderr|
|--------|------:|------|----:|---|-----:| |--------|------:|------|----:|---|-----:|
......
{
"results": {
"anagrams1": {
"acc": 0.0,
"acc_stderr": 0.0
},
"anagrams2": {
"acc": 0.0,
"acc_stderr": 0.0
},
"cycle_letters": {
"acc": 0.0,
"acc_stderr": 0.0
},
"random_insertion": {
"acc": 0.0,
"acc_stderr": 0.0
},
"reversed_words": {
"acc": 0.0,
"acc_stderr": 0.0
}
},
"versions": {
"anagrams1": 0,
"anagrams2": 0,
"cycle_letters": 0,
"random_insertion": 0,
"reversed_words": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=/gaueko1/hizkuntza-ereduak/LLaMA/lm/huggingface/7B",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
This diff is collapsed.
{
"results": {
"anli_r1": {
"acc": 0.332,
"acc_stderr": 0.014899597242811485
},
"anli_r2": {
"acc": 0.336,
"acc_stderr": 0.014944140233795027
},
"anli_r3": {
"acc": 0.345,
"acc_stderr": 0.013728421539454881
}
},
"versions": {
"anli_r1": 0,
"anli_r2": 0,
"anli_r3": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"blimp_adjunct_island": {
"acc": 0.878,
"acc_stderr": 0.010354864712936724
},
"blimp_anaphor_gender_agreement": {
"acc": 0.995,
"acc_stderr": 0.002231586874844882
},
"blimp_anaphor_number_agreement": {
"acc": 0.995,
"acc_stderr": 0.0022315868748448817
},
"blimp_animate_subject_passive": {
"acc": 0.775,
"acc_stderr": 0.013211720158614753
},
"blimp_animate_subject_trans": {
"acc": 0.884,
"acc_stderr": 0.010131468138756997
},
"blimp_causative": {
"acc": 0.747,
"acc_stderr": 0.01375427861358708
},
"blimp_complex_NP_island": {
"acc": 0.528,
"acc_stderr": 0.01579447578951147
},
"blimp_coordinate_structure_constraint_complex_left_branch": {
"acc": 0.779,
"acc_stderr": 0.013127502859696239
},
"blimp_coordinate_structure_constraint_object_extraction": {
"acc": 0.848,
"acc_stderr": 0.011358918303475293
},
"blimp_determiner_noun_agreement_1": {
"acc": 0.992,
"acc_stderr": 0.002818500300504507
},
"blimp_determiner_noun_agreement_2": {
"acc": 0.974,
"acc_stderr": 0.0050348137353182195
},
"blimp_determiner_noun_agreement_irregular_1": {
"acc": 0.936,
"acc_stderr": 0.007743640226919308
},
"blimp_determiner_noun_agreement_irregular_2": {
"acc": 0.928,
"acc_stderr": 0.008178195576218681
},
"blimp_determiner_noun_agreement_with_adj_2": {
"acc": 0.935,
"acc_stderr": 0.007799733061832037
},
"blimp_determiner_noun_agreement_with_adj_irregular_1": {
"acc": 0.883,
"acc_stderr": 0.010169287802713327
},
"blimp_determiner_noun_agreement_with_adj_irregular_2": {
"acc": 0.919,
"acc_stderr": 0.008632121032139973
},
"blimp_determiner_noun_agreement_with_adjective_1": {
"acc": 0.972,
"acc_stderr": 0.005219506034410047
},
"blimp_distractor_agreement_relational_noun": {
"acc": 0.889,
"acc_stderr": 0.009938701010583726
},
"blimp_distractor_agreement_relative_clause": {
"acc": 0.743,
"acc_stderr": 0.013825416526895042
},
"blimp_drop_argument": {
"acc": 0.787,
"acc_stderr": 0.012953717566737234
},
"blimp_ellipsis_n_bar_1": {
"acc": 0.79,
"acc_stderr": 0.012886662332274522
},
"blimp_ellipsis_n_bar_2": {
"acc": 0.928,
"acc_stderr": 0.008178195576218681
},
"blimp_existential_there_object_raising": {
"acc": 0.834,
"acc_stderr": 0.01177211037081219
},
"blimp_existential_there_quantifiers_1": {
"acc": 0.988,
"acc_stderr": 0.003444977194099841
},
"blimp_existential_there_quantifiers_2": {
"acc": 0.27,
"acc_stderr": 0.014046255632633908
},
"blimp_existential_there_subject_raising": {
"acc": 0.888,
"acc_stderr": 0.00997775303139724
},
"blimp_expletive_it_object_raising": {
"acc": 0.8,
"acc_stderr": 0.012655439943366646
},
"blimp_inchoative": {
"acc": 0.673,
"acc_stderr": 0.01484221315341125
},
"blimp_intransitive": {
"acc": 0.832,
"acc_stderr": 0.01182860583145426
},
"blimp_irregular_past_participle_adjectives": {
"acc": 0.972,
"acc_stderr": 0.005219506034410053
},
"blimp_irregular_past_participle_verbs": {
"acc": 0.885,
"acc_stderr": 0.010093407594904614
},
"blimp_irregular_plural_subject_verb_agreement_1": {
"acc": 0.92,
"acc_stderr": 0.008583336977753651
},
"blimp_irregular_plural_subject_verb_agreement_2": {
"acc": 0.908,
"acc_stderr": 0.009144376393151112
},
"blimp_left_branch_island_echo_question": {
"acc": 0.432,
"acc_stderr": 0.01567232023733621
},
"blimp_left_branch_island_simple_question": {
"acc": 0.897,
"acc_stderr": 0.009616833339695798
},
"blimp_matrix_question_npi_licensor_present": {
"acc": 0.705,
"acc_stderr": 0.014428554438445514
},
"blimp_npi_present_1": {
"acc": 0.579,
"acc_stderr": 0.015620595475301318
},
"blimp_npi_present_2": {
"acc": 0.688,
"acc_stderr": 0.01465847437050901
},
"blimp_only_npi_licensor_present": {
"acc": 0.916,
"acc_stderr": 0.008776162089491096
},
"blimp_only_npi_scope": {
"acc": 0.732,
"acc_stderr": 0.014013292702729491
},
"blimp_passive_1": {
"acc": 0.887,
"acc_stderr": 0.010016552866696863
},
"blimp_passive_2": {
"acc": 0.895,
"acc_stderr": 0.009698921026024947
},
"blimp_principle_A_c_command": {
"acc": 0.75,
"acc_stderr": 0.013699915608779773
},
"blimp_principle_A_case_1": {
"acc": 1.0,
"acc_stderr": 0.0
},
"blimp_principle_A_case_2": {
"acc": 0.94,
"acc_stderr": 0.007513751157474928
},
"blimp_principle_A_domain_1": {
"acc": 0.997,
"acc_stderr": 0.0017303161543469417
},
"blimp_principle_A_domain_2": {
"acc": 0.828,
"acc_stderr": 0.011939788882495321
},
"blimp_principle_A_domain_3": {
"acc": 0.761,
"acc_stderr": 0.013493000446937591
},
"blimp_principle_A_reconstruction": {
"acc": 0.41,
"acc_stderr": 0.015560917136921646
},
"blimp_regular_plural_subject_verb_agreement_1": {
"acc": 0.971,
"acc_stderr": 0.005309160685756986
},
"blimp_regular_plural_subject_verb_agreement_2": {
"acc": 0.907,
"acc_stderr": 0.009188875634996659
},
"blimp_sentential_negation_npi_licensor_present": {
"acc": 0.989,
"acc_stderr": 0.0032999833166078166
},
"blimp_sentential_negation_npi_scope": {
"acc": 0.733,
"acc_stderr": 0.013996674851796264
},
"blimp_sentential_subject_island": {
"acc": 0.399,
"acc_stderr": 0.015493193313162906
},
"blimp_superlative_quantifiers_1": {
"acc": 0.822,
"acc_stderr": 0.012102167676183596
},
"blimp_superlative_quantifiers_2": {
"acc": 0.897,
"acc_stderr": 0.009616833339695796
},
"blimp_tough_vs_raising_1": {
"acc": 0.69,
"acc_stderr": 0.014632638658632898
},
"blimp_tough_vs_raising_2": {
"acc": 0.829,
"acc_stderr": 0.011912216456264606
},
"blimp_transitive": {
"acc": 0.872,
"acc_stderr": 0.010570133761108658
},
"blimp_wh_island": {
"acc": 0.813,
"acc_stderr": 0.012336254828074126
},
"blimp_wh_questions_object_gap": {
"acc": 0.763,
"acc_stderr": 0.01345407046257794
},
"blimp_wh_questions_subject_gap": {
"acc": 0.89,
"acc_stderr": 0.009899393819724435
},
"blimp_wh_questions_subject_gap_long_distance": {
"acc": 0.893,
"acc_stderr": 0.009779910359847169
},
"blimp_wh_vs_that_no_gap": {
"acc": 0.946,
"acc_stderr": 0.007150883521295428
},
"blimp_wh_vs_that_no_gap_long_distance": {
"acc": 0.951,
"acc_stderr": 0.006829761756140921
},
"blimp_wh_vs_that_with_gap": {
"acc": 0.321,
"acc_stderr": 0.01477082181793466
},
"blimp_wh_vs_that_with_gap_long_distance": {
"acc": 0.292,
"acc_stderr": 0.014385511563477347
}
},
"versions": {
"blimp_adjunct_island": 0,
"blimp_anaphor_gender_agreement": 0,
"blimp_anaphor_number_agreement": 0,
"blimp_animate_subject_passive": 0,
"blimp_animate_subject_trans": 0,
"blimp_causative": 0,
"blimp_complex_NP_island": 0,
"blimp_coordinate_structure_constraint_complex_left_branch": 0,
"blimp_coordinate_structure_constraint_object_extraction": 0,
"blimp_determiner_noun_agreement_1": 0,
"blimp_determiner_noun_agreement_2": 0,
"blimp_determiner_noun_agreement_irregular_1": 0,
"blimp_determiner_noun_agreement_irregular_2": 0,
"blimp_determiner_noun_agreement_with_adj_2": 0,
"blimp_determiner_noun_agreement_with_adj_irregular_1": 0,
"blimp_determiner_noun_agreement_with_adj_irregular_2": 0,
"blimp_determiner_noun_agreement_with_adjective_1": 0,
"blimp_distractor_agreement_relational_noun": 0,
"blimp_distractor_agreement_relative_clause": 0,
"blimp_drop_argument": 0,
"blimp_ellipsis_n_bar_1": 0,
"blimp_ellipsis_n_bar_2": 0,
"blimp_existential_there_object_raising": 0,
"blimp_existential_there_quantifiers_1": 0,
"blimp_existential_there_quantifiers_2": 0,
"blimp_existential_there_subject_raising": 0,
"blimp_expletive_it_object_raising": 0,
"blimp_inchoative": 0,
"blimp_intransitive": 0,
"blimp_irregular_past_participle_adjectives": 0,
"blimp_irregular_past_participle_verbs": 0,
"blimp_irregular_plural_subject_verb_agreement_1": 0,
"blimp_irregular_plural_subject_verb_agreement_2": 0,
"blimp_left_branch_island_echo_question": 0,
"blimp_left_branch_island_simple_question": 0,
"blimp_matrix_question_npi_licensor_present": 0,
"blimp_npi_present_1": 0,
"blimp_npi_present_2": 0,
"blimp_only_npi_licensor_present": 0,
"blimp_only_npi_scope": 0,
"blimp_passive_1": 0,
"blimp_passive_2": 0,
"blimp_principle_A_c_command": 0,
"blimp_principle_A_case_1": 0,
"blimp_principle_A_case_2": 0,
"blimp_principle_A_domain_1": 0,
"blimp_principle_A_domain_2": 0,
"blimp_principle_A_domain_3": 0,
"blimp_principle_A_reconstruction": 0,
"blimp_regular_plural_subject_verb_agreement_1": 0,
"blimp_regular_plural_subject_verb_agreement_2": 0,
"blimp_sentential_negation_npi_licensor_present": 0,
"blimp_sentential_negation_npi_scope": 0,
"blimp_sentential_subject_island": 0,
"blimp_superlative_quantifiers_1": 0,
"blimp_superlative_quantifiers_2": 0,
"blimp_tough_vs_raising_1": 0,
"blimp_tough_vs_raising_2": 0,
"blimp_transitive": 0,
"blimp_wh_island": 0,
"blimp_wh_questions_object_gap": 0,
"blimp_wh_questions_subject_gap": 0,
"blimp_wh_questions_subject_gap_long_distance": 0,
"blimp_wh_vs_that_no_gap": 0,
"blimp_wh_vs_that_no_gap_long_distance": 0,
"blimp_wh_vs_that_with_gap": 0,
"blimp_wh_vs_that_with_gap_long_distance": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"crows_pairs_english_age": {
"likelihood_difference": 4.1510989010989015,
"likelihood_difference_stderr": 0.38322042294913006,
"pct_stereotype": 0.7362637362637363,
"pct_stereotype_stderr": 0.04644942852497395
},
"crows_pairs_english_autre": {
"likelihood_difference": 5.056818181818182,
"likelihood_difference_stderr": 1.7702853227564594,
"pct_stereotype": 0.7272727272727273,
"pct_stereotype_stderr": 0.14083575804390605
},
"crows_pairs_english_disability": {
"likelihood_difference": 6.019230769230769,
"likelihood_difference_stderr": 0.6330801651940536,
"pct_stereotype": 0.7692307692307693,
"pct_stereotype_stderr": 0.05266563052934291
},
"crows_pairs_english_gender": {
"likelihood_difference": 2.682421875,
"likelihood_difference_stderr": 0.17011984636779479,
"pct_stereotype": 0.6375,
"pct_stereotype_stderr": 0.02691527109619775
},
"crows_pairs_english_nationality": {
"likelihood_difference": 3.498263888888889,
"likelihood_difference_stderr": 0.21512426671182808,
"pct_stereotype": 0.6157407407407407,
"pct_stereotype_stderr": 0.03317354514310742
},
"crows_pairs_english_physical_appearance": {
"likelihood_difference": 3.7378472222222223,
"likelihood_difference_stderr": 0.3384578916047944,
"pct_stereotype": 0.7222222222222222,
"pct_stereotype_stderr": 0.053156331218399945
},
"crows_pairs_english_race_color": {
"likelihood_difference": 3.3619586614173227,
"likelihood_difference_stderr": 0.14103384370541044,
"pct_stereotype": 0.5728346456692913,
"pct_stereotype_stderr": 0.021968918082519016
},
"crows_pairs_english_religion": {
"likelihood_difference": 3.6644144144144146,
"likelihood_difference_stderr": 0.33863382204528913,
"pct_stereotype": 0.7297297297297297,
"pct_stereotype_stderr": 0.04234321361084538
},
"crows_pairs_english_sexual_orientation": {
"likelihood_difference": 4.630376344086022,
"likelihood_difference_stderr": 0.4574502494208299,
"pct_stereotype": 0.8279569892473119,
"pct_stereotype_stderr": 0.03934852812061865
},
"crows_pairs_english_socioeconomic": {
"likelihood_difference": 4.065131578947368,
"likelihood_difference_stderr": 0.23523235884669547,
"pct_stereotype": 0.6789473684210526,
"pct_stereotype_stderr": 0.03396059335824887
},
"crows_pairs_french_age": {
"likelihood_difference": 3.609722222222222,
"likelihood_difference_stderr": 0.3615396777446631,
"pct_stereotype": 0.4222222222222222,
"pct_stereotype_stderr": 0.05235473399540658
},
"crows_pairs_french_autre": {
"likelihood_difference": 2.6923076923076925,
"likelihood_difference_stderr": 0.9229767573858816,
"pct_stereotype": 0.6153846153846154,
"pct_stereotype_stderr": 0.14044168141158106
},
"crows_pairs_french_disability": {
"likelihood_difference": 4.958333333333333,
"likelihood_difference_stderr": 0.42689940749326905,
"pct_stereotype": 0.6363636363636364,
"pct_stereotype_stderr": 0.05966637484671758
},
"crows_pairs_french_gender": {
"likelihood_difference": 3.213785046728972,
"likelihood_difference_stderr": 0.17588386956758606,
"pct_stereotype": 0.5109034267912772,
"pct_stereotype_stderr": 0.027944203070818633
},
"crows_pairs_french_nationality": {
"likelihood_difference": 3.883399209486166,
"likelihood_difference_stderr": 0.21844567559439967,
"pct_stereotype": 0.3438735177865613,
"pct_stereotype_stderr": 0.029922155720849428
},
"crows_pairs_french_physical_appearance": {
"likelihood_difference": 3.2274305555555554,
"likelihood_difference_stderr": 0.4328667471914375,
"pct_stereotype": 0.5972222222222222,
"pct_stereotype_stderr": 0.05820650942569533
},
"crows_pairs_french_race_color": {
"likelihood_difference": 3.161413043478261,
"likelihood_difference_stderr": 0.16557903974411925,
"pct_stereotype": 0.4369565217391304,
"pct_stereotype_stderr": 0.023151745316873387
},
"crows_pairs_french_religion": {
"likelihood_difference": 3.5673913043478263,
"likelihood_difference_stderr": 0.3368331015818195,
"pct_stereotype": 0.6260869565217392,
"pct_stereotype_stderr": 0.045315858286449635
},
"crows_pairs_french_sexual_orientation": {
"likelihood_difference": 4.791208791208791,
"likelihood_difference_stderr": 0.4009539855629619,
"pct_stereotype": 0.7802197802197802,
"pct_stereotype_stderr": 0.043649726328985346
},
"crows_pairs_french_socioeconomic": {
"likelihood_difference": 3.9939413265306123,
"likelihood_difference_stderr": 0.263093158126228,
"pct_stereotype": 0.6581632653061225,
"pct_stereotype_stderr": 0.033967132039868675
},
"ethics_cm": {
"acc": 0.5459459459459459,
"acc_stderr": 0.007988936899457039
},
"ethics_deontology": {
"acc": 0.5025027808676307,
"acc_stderr": 0.008339021933755771,
"em": 0.004449388209121246
},
"ethics_justice": {
"acc": 0.5196005917159763,
"acc_stderr": 0.009609770755397009,
"em": 0.011834319526627219
},
"ethics_utilitarianism": {
"acc": 0.5748752079866889,
"acc_stderr": 0.007130302336230959
},
"ethics_utilitarianism_original": {
"acc": 0.9956322795341098,
"acc_stderr": 0.000951129914345755
},
"ethics_virtue": {
"acc": 0.8040201005025126,
"acc_stderr": 0.005628417801676332,
"em": 0.12562814070351758
},
"toxigen": {
"acc": 0.4319148936170213,
"acc_stderr": 0.016164899004911828,
"acc_norm": 0.4319148936170213,
"acc_norm_stderr": 0.016164899004911828
}
},
"versions": {
"crows_pairs_english_age": 0,
"crows_pairs_english_autre": 0,
"crows_pairs_english_disability": 0,
"crows_pairs_english_gender": 0,
"crows_pairs_english_nationality": 0,
"crows_pairs_english_physical_appearance": 0,
"crows_pairs_english_race_color": 0,
"crows_pairs_english_religion": 0,
"crows_pairs_english_sexual_orientation": 0,
"crows_pairs_english_socioeconomic": 0,
"crows_pairs_french_age": 0,
"crows_pairs_french_autre": 0,
"crows_pairs_french_disability": 0,
"crows_pairs_french_gender": 0,
"crows_pairs_french_nationality": 0,
"crows_pairs_french_physical_appearance": 0,
"crows_pairs_french_race_color": 0,
"crows_pairs_french_religion": 0,
"crows_pairs_french_sexual_orientation": 0,
"crows_pairs_french_socioeconomic": 0,
"ethics_cm": 0,
"ethics_deontology": 0,
"ethics_justice": 0,
"ethics_utilitarianism": 0,
"ethics_utilitarianism_original": 0,
"ethics_virtue": 0,
"toxigen": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"hendrycksTest-abstract_algebra": {
"acc": 0.18,
"acc_stderr": 0.03861229196653695,
"acc_norm": 0.21,
"acc_norm_stderr": 0.040936018074033256
},
"hendrycksTest-anatomy": {
"acc": 0.3851851851851852,
"acc_stderr": 0.042039210401562783,
"acc_norm": 0.37777777777777777,
"acc_norm_stderr": 0.04188307537595853
},
"hendrycksTest-astronomy": {
"acc": 0.39473684210526316,
"acc_stderr": 0.039777499346220734,
"acc_norm": 0.42105263157894735,
"acc_norm_stderr": 0.04017901275981748
},
"hendrycksTest-business_ethics": {
"acc": 0.49,
"acc_stderr": 0.05024183937956911,
"acc_norm": 0.48,
"acc_norm_stderr": 0.050211673156867795
},
"hendrycksTest-clinical_knowledge": {
"acc": 0.3320754716981132,
"acc_stderr": 0.028985455652334388,
"acc_norm": 0.37735849056603776,
"acc_norm_stderr": 0.029832808114796005
},
"hendrycksTest-college_biology": {
"acc": 0.3819444444444444,
"acc_stderr": 0.040629907841466674,
"acc_norm": 0.3541666666666667,
"acc_norm_stderr": 0.039994111357535424
},
"hendrycksTest-college_chemistry": {
"acc": 0.39,
"acc_stderr": 0.04902071300001974,
"acc_norm": 0.41,
"acc_norm_stderr": 0.049431107042371025
},
"hendrycksTest-college_computer_science": {
"acc": 0.34,
"acc_stderr": 0.04760952285695235,
"acc_norm": 0.32,
"acc_norm_stderr": 0.046882617226215034
},
"hendrycksTest-college_mathematics": {
"acc": 0.27,
"acc_stderr": 0.044619604333847394,
"acc_norm": 0.33,
"acc_norm_stderr": 0.047258156262526045
},
"hendrycksTest-college_medicine": {
"acc": 0.36416184971098264,
"acc_stderr": 0.03669072477416907,
"acc_norm": 0.3468208092485549,
"acc_norm_stderr": 0.036291466701596636
},
"hendrycksTest-college_physics": {
"acc": 0.30392156862745096,
"acc_stderr": 0.045766654032077615,
"acc_norm": 0.3333333333333333,
"acc_norm_stderr": 0.04690650298201943
},
"hendrycksTest-computer_security": {
"acc": 0.41,
"acc_stderr": 0.04943110704237102,
"acc_norm": 0.41,
"acc_norm_stderr": 0.049431107042371025
},
"hendrycksTest-conceptual_physics": {
"acc": 0.3276595744680851,
"acc_stderr": 0.030683020843231015,
"acc_norm": 0.2553191489361702,
"acc_norm_stderr": 0.02850485647051418
},
"hendrycksTest-econometrics": {
"acc": 0.2719298245614035,
"acc_stderr": 0.04185774424022056,
"acc_norm": 0.23684210526315788,
"acc_norm_stderr": 0.039994238792813365
},
"hendrycksTest-electrical_engineering": {
"acc": 0.36551724137931035,
"acc_stderr": 0.040131241954243856,
"acc_norm": 0.33793103448275863,
"acc_norm_stderr": 0.03941707632064889
},
"hendrycksTest-elementary_mathematics": {
"acc": 0.29894179894179895,
"acc_stderr": 0.023577604791655802,
"acc_norm": 0.28835978835978837,
"acc_norm_stderr": 0.023330654054535892
},
"hendrycksTest-formal_logic": {
"acc": 0.30952380952380953,
"acc_stderr": 0.04134913018303316,
"acc_norm": 0.2857142857142857,
"acc_norm_stderr": 0.040406101782088394
},
"hendrycksTest-global_facts": {
"acc": 0.35,
"acc_stderr": 0.047937248544110196,
"acc_norm": 0.33,
"acc_norm_stderr": 0.047258156262526045
},
"hendrycksTest-high_school_biology": {
"acc": 0.36451612903225805,
"acc_stderr": 0.02737987122994325,
"acc_norm": 0.3903225806451613,
"acc_norm_stderr": 0.027751256636969583
},
"hendrycksTest-high_school_chemistry": {
"acc": 0.21182266009852216,
"acc_stderr": 0.02874898368994106,
"acc_norm": 0.21674876847290642,
"acc_norm_stderr": 0.028990331252516235
},
"hendrycksTest-high_school_computer_science": {
"acc": 0.43,
"acc_stderr": 0.04975698519562428,
"acc_norm": 0.41,
"acc_norm_stderr": 0.049431107042371025
},
"hendrycksTest-high_school_european_history": {
"acc": 0.38181818181818183,
"acc_stderr": 0.03793713171165635,
"acc_norm": 0.37575757575757573,
"acc_norm_stderr": 0.03781887353205983
},
"hendrycksTest-high_school_geography": {
"acc": 0.3838383838383838,
"acc_stderr": 0.03464881675016339,
"acc_norm": 0.40404040404040403,
"acc_norm_stderr": 0.03496130972056128
},
"hendrycksTest-high_school_government_and_politics": {
"acc": 0.41450777202072536,
"acc_stderr": 0.03555300319557673,
"acc_norm": 0.41450777202072536,
"acc_norm_stderr": 0.03555300319557672
},
"hendrycksTest-high_school_macroeconomics": {
"acc": 0.3487179487179487,
"acc_stderr": 0.024162780284017717,
"acc_norm": 0.29743589743589743,
"acc_norm_stderr": 0.02317740813146594
},
"hendrycksTest-high_school_mathematics": {
"acc": 0.29259259259259257,
"acc_stderr": 0.027738969632176088,
"acc_norm": 0.3037037037037037,
"acc_norm_stderr": 0.02803792996911499
},
"hendrycksTest-high_school_microeconomics": {
"acc": 0.33613445378151263,
"acc_stderr": 0.030684737115135353,
"acc_norm": 0.3697478991596639,
"acc_norm_stderr": 0.03135709599613591
},
"hendrycksTest-high_school_physics": {
"acc": 0.2781456953642384,
"acc_stderr": 0.03658603262763743,
"acc_norm": 0.2781456953642384,
"acc_norm_stderr": 0.03658603262763743
},
"hendrycksTest-high_school_psychology": {
"acc": 0.46972477064220186,
"acc_stderr": 0.021397988604936965,
"acc_norm": 0.44587155963302755,
"acc_norm_stderr": 0.02131133500970858
},
"hendrycksTest-high_school_statistics": {
"acc": 0.3287037037037037,
"acc_stderr": 0.03203614084670058,
"acc_norm": 0.32407407407407407,
"acc_norm_stderr": 0.03191923445686185
},
"hendrycksTest-high_school_us_history": {
"acc": 0.3431372549019608,
"acc_stderr": 0.03332139944668085,
"acc_norm": 0.3137254901960784,
"acc_norm_stderr": 0.032566854844603886
},
"hendrycksTest-high_school_world_history": {
"acc": 0.29535864978902954,
"acc_stderr": 0.029696338713422876,
"acc_norm": 0.2869198312236287,
"acc_norm_stderr": 0.02944377302259469
},
"hendrycksTest-human_aging": {
"acc": 0.336322869955157,
"acc_stderr": 0.031708824268455,
"acc_norm": 0.3273542600896861,
"acc_norm_stderr": 0.031493846709941306
},
"hendrycksTest-human_sexuality": {
"acc": 0.2748091603053435,
"acc_stderr": 0.03915345408847836,
"acc_norm": 0.3282442748091603,
"acc_norm_stderr": 0.041184385658062976
},
"hendrycksTest-international_law": {
"acc": 0.371900826446281,
"acc_stderr": 0.04412015806624504,
"acc_norm": 0.49586776859504134,
"acc_norm_stderr": 0.045641987674327526
},
"hendrycksTest-jurisprudence": {
"acc": 0.3425925925925926,
"acc_stderr": 0.045879047413018105,
"acc_norm": 0.39814814814814814,
"acc_norm_stderr": 0.04732332615978814
},
"hendrycksTest-logical_fallacies": {
"acc": 0.3803680981595092,
"acc_stderr": 0.038142698932618374,
"acc_norm": 0.36809815950920244,
"acc_norm_stderr": 0.03789213935838395
},
"hendrycksTest-machine_learning": {
"acc": 0.26785714285714285,
"acc_stderr": 0.04203277291467763,
"acc_norm": 0.24107142857142858,
"acc_norm_stderr": 0.04059867246952686
},
"hendrycksTest-management": {
"acc": 0.42718446601941745,
"acc_stderr": 0.04897957737781169,
"acc_norm": 0.39805825242718446,
"acc_norm_stderr": 0.0484674825397724
},
"hendrycksTest-marketing": {
"acc": 0.5512820512820513,
"acc_stderr": 0.032583346493868806,
"acc_norm": 0.5512820512820513,
"acc_norm_stderr": 0.032583346493868806
},
"hendrycksTest-medical_genetics": {
"acc": 0.39,
"acc_stderr": 0.04902071300001975,
"acc_norm": 0.38,
"acc_norm_stderr": 0.04878317312145633
},
"hendrycksTest-miscellaneous": {
"acc": 0.5555555555555556,
"acc_stderr": 0.017769250583533246,
"acc_norm": 0.5568326947637292,
"acc_norm_stderr": 0.01776408503534841
},
"hendrycksTest-moral_disputes": {
"acc": 0.3208092485549133,
"acc_stderr": 0.025131000233647904,
"acc_norm": 0.30057803468208094,
"acc_norm_stderr": 0.024685316867257796
},
"hendrycksTest-moral_scenarios": {
"acc": 0.26033519553072626,
"acc_stderr": 0.014676252009319483,
"acc_norm": 0.27262569832402234,
"acc_norm_stderr": 0.014893391735249614
},
"hendrycksTest-nutrition": {
"acc": 0.3431372549019608,
"acc_stderr": 0.027184498909941616,
"acc_norm": 0.4019607843137255,
"acc_norm_stderr": 0.02807415894760066
},
"hendrycksTest-philosophy": {
"acc": 0.3762057877813505,
"acc_stderr": 0.027513925683549427,
"acc_norm": 0.36977491961414793,
"acc_norm_stderr": 0.027417996705630998
},
"hendrycksTest-prehistory": {
"acc": 0.33641975308641975,
"acc_stderr": 0.02628973494595293,
"acc_norm": 0.3055555555555556,
"acc_norm_stderr": 0.025630824975621344
},
"hendrycksTest-professional_accounting": {
"acc": 0.3049645390070922,
"acc_stderr": 0.027464708442022135,
"acc_norm": 0.2907801418439716,
"acc_norm_stderr": 0.027090664368353178
},
"hendrycksTest-professional_law": {
"acc": 0.25945241199478486,
"acc_stderr": 0.011195262076350299,
"acc_norm": 0.2842242503259452,
"acc_norm_stderr": 0.011519880596516074
},
"hendrycksTest-professional_medicine": {
"acc": 0.29411764705882354,
"acc_stderr": 0.027678468642144703,
"acc_norm": 0.3161764705882353,
"acc_norm_stderr": 0.02824568739146291
},
"hendrycksTest-professional_psychology": {
"acc": 0.315359477124183,
"acc_stderr": 0.018798086284886883,
"acc_norm": 0.3022875816993464,
"acc_norm_stderr": 0.01857923271111388
},
"hendrycksTest-public_relations": {
"acc": 0.41818181818181815,
"acc_stderr": 0.04724577405731571,
"acc_norm": 0.42727272727272725,
"acc_norm_stderr": 0.04738198703545483
},
"hendrycksTest-security_studies": {
"acc": 0.2816326530612245,
"acc_stderr": 0.028795185574291282,
"acc_norm": 0.24081632653061225,
"acc_norm_stderr": 0.027372942201788163
},
"hendrycksTest-sociology": {
"acc": 0.34328358208955223,
"acc_stderr": 0.03357379665433431,
"acc_norm": 0.3681592039800995,
"acc_norm_stderr": 0.034104105654953025
},
"hendrycksTest-us_foreign_policy": {
"acc": 0.38,
"acc_stderr": 0.04878317312145632,
"acc_norm": 0.39,
"acc_norm_stderr": 0.04902071300001975
},
"hendrycksTest-virology": {
"acc": 0.3253012048192771,
"acc_stderr": 0.03647168523683229,
"acc_norm": 0.3253012048192771,
"acc_norm_stderr": 0.03647168523683227
},
"hendrycksTest-world_religions": {
"acc": 0.543859649122807,
"acc_stderr": 0.03820042586602966,
"acc_norm": 0.5789473684210527,
"acc_norm_stderr": 0.03786720706234214
}
},
"versions": {
"hendrycksTest-abstract_algebra": 0,
"hendrycksTest-anatomy": 0,
"hendrycksTest-astronomy": 0,
"hendrycksTest-business_ethics": 0,
"hendrycksTest-clinical_knowledge": 0,
"hendrycksTest-college_biology": 0,
"hendrycksTest-college_chemistry": 0,
"hendrycksTest-college_computer_science": 0,
"hendrycksTest-college_mathematics": 0,
"hendrycksTest-college_medicine": 0,
"hendrycksTest-college_physics": 0,
"hendrycksTest-computer_security": 0,
"hendrycksTest-conceptual_physics": 0,
"hendrycksTest-econometrics": 0,
"hendrycksTest-electrical_engineering": 0,
"hendrycksTest-elementary_mathematics": 0,
"hendrycksTest-formal_logic": 0,
"hendrycksTest-global_facts": 0,
"hendrycksTest-high_school_biology": 0,
"hendrycksTest-high_school_chemistry": 0,
"hendrycksTest-high_school_computer_science": 0,
"hendrycksTest-high_school_european_history": 0,
"hendrycksTest-high_school_geography": 0,
"hendrycksTest-high_school_government_and_politics": 0,
"hendrycksTest-high_school_macroeconomics": 0,
"hendrycksTest-high_school_mathematics": 0,
"hendrycksTest-high_school_microeconomics": 0,
"hendrycksTest-high_school_physics": 0,
"hendrycksTest-high_school_psychology": 0,
"hendrycksTest-high_school_statistics": 0,
"hendrycksTest-high_school_us_history": 0,
"hendrycksTest-high_school_world_history": 0,
"hendrycksTest-human_aging": 0,
"hendrycksTest-human_sexuality": 0,
"hendrycksTest-international_law": 0,
"hendrycksTest-jurisprudence": 0,
"hendrycksTest-logical_fallacies": 0,
"hendrycksTest-machine_learning": 0,
"hendrycksTest-management": 0,
"hendrycksTest-marketing": 0,
"hendrycksTest-medical_genetics": 0,
"hendrycksTest-miscellaneous": 0,
"hendrycksTest-moral_disputes": 0,
"hendrycksTest-moral_scenarios": 0,
"hendrycksTest-nutrition": 0,
"hendrycksTest-philosophy": 0,
"hendrycksTest-prehistory": 0,
"hendrycksTest-professional_accounting": 0,
"hendrycksTest-professional_law": 0,
"hendrycksTest-professional_medicine": 0,
"hendrycksTest-professional_psychology": 0,
"hendrycksTest-public_relations": 0,
"hendrycksTest-security_studies": 0,
"hendrycksTest-sociology": 0,
"hendrycksTest-us_foreign_policy": 0,
"hendrycksTest-virology": 0,
"hendrycksTest-world_religions": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True,dtype=bfloat16",
"num_fewshot": 5,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"pawsx_de": {
"acc": 0.614,
"acc_stderr": 0.010888584877766427
},
"pawsx_en": {
"acc": 0.7035,
"acc_stderr": 0.010214991337441791
},
"pawsx_es": {
"acc": 0.6495,
"acc_stderr": 0.01067154233969731
},
"pawsx_fr": {
"acc": 0.6285,
"acc_stderr": 0.01080751017293364
},
"pawsx_ja": {
"acc": 0.493,
"acc_stderr": 0.011182040020027768
},
"pawsx_ko": {
"acc": 0.5365,
"acc_stderr": 0.011153298751334327
},
"pawsx_zh": {
"acc": 0.5625,
"acc_stderr": 0.011095423796079503
}
},
"versions": {
"pawsx_de": 0,
"pawsx_en": 0,
"pawsx_es": 0,
"pawsx_fr": 0,
"pawsx_ja": 0,
"pawsx_ko": 0,
"pawsx_zh": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True,dtype=bfloat16",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"coqa": {
"f1": 0.7650867255895625,
"f1_stderr": 0.01481717694356494,
"em": 0.6301666666666667,
"em_stderr": 0.018680205213012713
},
"drop": {
"em": 0.03429110738255033,
"em_stderr": 0.0018636035184959787,
"f1": 0.1338569630872482,
"f1_stderr": 0.0025165760673094154
},
"race": {
"acc": 0.3866028708133971,
"acc_stderr": 0.01507138477304713
}
},
"versions": {
"coqa": 1,
"drop": 1,
"race": 1
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"anagrams1": {
"acc": 0.0,
"acc_stderr": 0.0
},
"anagrams2": {
"acc": 0.0001,
"acc_stderr": 0.0001000000000000119
},
"cycle_letters": {
"acc": 0.0,
"acc_stderr": 0.0
},
"random_insertion": {
"acc": 0.0004,
"acc_stderr": 0.00019996999474889126
},
"reversed_words": {
"acc": 0.0,
"acc_stderr": 0.0
}
},
"versions": {
"anagrams1": 0,
"anagrams2": 0,
"cycle_letters": 0,
"random_insertion": 0,
"reversed_words": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"xcopa_et": {
"acc": 0.474,
"acc_stderr": 0.02235279165091416
},
"xcopa_ht": {
"acc": 0.498,
"acc_stderr": 0.02238289498648353
},
"xcopa_id": {
"acc": 0.568,
"acc_stderr": 0.02217510926561316
},
"xcopa_it": {
"acc": 0.594,
"acc_stderr": 0.02198396209008634
},
"xcopa_qu": {
"acc": 0.484,
"acc_stderr": 0.0223716109825804
},
"xcopa_sw": {
"acc": 0.516,
"acc_stderr": 0.0223716109825804
},
"xcopa_ta": {
"acc": 0.54,
"acc_stderr": 0.022311333245289663
},
"xcopa_th": {
"acc": 0.542,
"acc_stderr": 0.02230396677426995
},
"xcopa_tr": {
"acc": 0.516,
"acc_stderr": 0.022371610982580396
},
"xcopa_vi": {
"acc": 0.536,
"acc_stderr": 0.022324981738385246
},
"xcopa_zh": {
"acc": 0.632,
"acc_stderr": 0.02158898256835354
}
},
"versions": {
"xcopa_et": 0,
"xcopa_ht": 0,
"xcopa_id": 0,
"xcopa_it": 0,
"xcopa_qu": 0,
"xcopa_sw": 0,
"xcopa_ta": 0,
"xcopa_th": 0,
"xcopa_tr": 0,
"xcopa_vi": 0,
"xcopa_zh": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True,dtype=bfloat16",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"xnli_ar": {
"acc": 0.3331337325349301,
"acc_stderr": 0.006659676975732312
},
"xnli_bg": {
"acc": 0.36826347305389223,
"acc_stderr": 0.00681509322031267
},
"xnli_de": {
"acc": 0.46447105788423154,
"acc_stderr": 0.007046854204317796
},
"xnli_el": {
"acc": 0.36187624750499003,
"acc_stderr": 0.006789799946197319
},
"xnli_en": {
"acc": 0.543313373253493,
"acc_stderr": 0.007038155029004149
},
"xnli_es": {
"acc": 0.4564870259481038,
"acc_stderr": 0.007037909229199955
},
"xnli_fr": {
"acc": 0.4880239520958084,
"acc_stderr": 0.007062685615595022
},
"xnli_hi": {
"acc": 0.3473053892215569,
"acc_stderr": 0.006727214239733754
},
"xnli_ru": {
"acc": 0.444311377245509,
"acc_stderr": 0.007020757195791274
},
"xnli_sw": {
"acc": 0.3341317365269461,
"acc_stderr": 0.006664652441694265
},
"xnli_th": {
"acc": 0.36127744510978044,
"acc_stderr": 0.006787362347422093
},
"xnli_tr": {
"acc": 0.37684630738522956,
"acc_stderr": 0.006847061089041557
},
"xnli_ur": {
"acc": 0.33632734530938124,
"acc_stderr": 0.006675480563072364
},
"xnli_vi": {
"acc": 0.37325349301397204,
"acc_stderr": 0.0068339592620100505
},
"xnli_zh": {
"acc": 0.3534930139720559,
"acc_stderr": 0.006754629196407293
}
},
"versions": {
"xnli_ar": 0,
"xnli_bg": 0,
"xnli_de": 0,
"xnli_el": 0,
"xnli_en": 0,
"xnli_es": 0,
"xnli_fr": 0,
"xnli_hi": 0,
"xnli_ru": 0,
"xnli_sw": 0,
"xnli_th": 0,
"xnli_tr": 0,
"xnli_ur": 0,
"xnli_vi": 0,
"xnli_zh": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True,dtype=bfloat16",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"xstory_cloze_ar": {
"acc": 0.48510919920582396,
"acc_stderr": 0.012861417842074004
},
"xstory_cloze_en": {
"acc": 0.7789543348775645,
"acc_stderr": 0.010678457581809239
},
"xstory_cloze_es": {
"acc": 0.6604897418927862,
"acc_stderr": 0.01218627614665943
},
"xstory_cloze_eu": {
"acc": 0.5109199205823958,
"acc_stderr": 0.012864056278255038
},
"xstory_cloze_hi": {
"acc": 0.5168762409000662,
"acc_stderr": 0.012859793919977606
},
"xstory_cloze_id": {
"acc": 0.5519523494374586,
"acc_stderr": 0.012797478885304742
},
"xstory_cloze_my": {
"acc": 0.48378557246856385,
"acc_stderr": 0.01286035780505586
},
"xstory_cloze_ru": {
"acc": 0.5724685638649901,
"acc_stderr": 0.012731259626982528
},
"xstory_cloze_sw": {
"acc": 0.4990072799470549,
"acc_stderr": 0.012867099955422935
},
"xstory_cloze_te": {
"acc": 0.5294506949040371,
"acc_stderr": 0.012844785490017004
},
"xstory_cloze_zh": {
"acc": 0.5956320317670417,
"acc_stderr": 0.012629580396570932
}
},
"versions": {
"xstory_cloze_ar": 0,
"xstory_cloze_en": 0,
"xstory_cloze_es": 0,
"xstory_cloze_eu": 0,
"xstory_cloze_hi": 0,
"xstory_cloze_id": 0,
"xstory_cloze_my": 0,
"xstory_cloze_ru": 0,
"xstory_cloze_sw": 0,
"xstory_cloze_te": 0,
"xstory_cloze_zh": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True,dtype=bfloat16",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
{
"results": {
"xwinograd_en": {
"acc": 0.8666666666666667,
"acc_stderr": 0.0070514325016347275
},
"xwinograd_fr": {
"acc": 0.6626506024096386,
"acc_stderr": 0.05221260262032129
},
"xwinograd_jp": {
"acc": 0.602711157455683,
"acc_stderr": 0.015809751560314552
},
"xwinograd_pt": {
"acc": 0.6692015209125475,
"acc_stderr": 0.02906762615931534
},
"xwinograd_ru": {
"acc": 0.6952380952380952,
"acc_stderr": 0.025976599352305375
},
"xwinograd_zh": {
"acc": 0.7162698412698413,
"acc_stderr": 0.02010051064884106
}
},
"versions": {
"xwinograd_en": 0,
"xwinograd_fr": 0,
"xwinograd_jp": 0,
"xwinograd_pt": 0,
"xwinograd_ru": 0,
"xwinograd_zh": 0
},
"config": {
"model": "hf-causal-experimental",
"model_args": "pretrained=mosaicml/mpt-7b,trust_remote_code=True,dtype=bfloat16",
"num_fewshot": 0,
"batch_size": "auto",
"device": "cuda",
"no_cache": true,
"limit": null,
"bootstrap_iters": 100000,
"description_dict": {}
}
}
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment