Codemind-study / m28 /benchmark_results.json
B4K2xx's picture
Add m28
f4ed4c1 verified
Raw
History Blame Contribute Delete
18.5 kB
{
"model_id": "m28",
"checkpoint": "results/m28/final.pt",
"trained_step": 15258,
"num_fewshot": 0,
"limit": null,
"results": {
"lambada_openai": {
"name": "lambada_openai",
"alias": "lambada_openai",
"sample_len": 5153,
"perplexity,none": 566.312332169294,
"perplexity_stderr,none": 30.16584819271405,
"acc,none": 0.17543178730836406,
"acc_stderr,none": 0.005298824200667555
},
"hellaswag": {
"name": "hellaswag",
"alias": "hellaswag",
"sample_len": 10042,
"acc,none": 0.27215694084843656,
"acc_stderr,none": 0.004441606665788188,
"acc_norm,none": 0.2787293367855009,
"acc_norm_stderr,none": 0.004474577054517292
},
"piqa": {
"name": "piqa",
"alias": "piqa",
"sample_len": 1838,
"acc,none": 0.6017410228509249,
"acc_stderr,none": 0.011421759697789035,
"acc_norm,none": 0.5919477693144722,
"acc_norm_stderr,none": 0.011466872778651143
},
"sciq": {
"name": "sciq",
"alias": "sciq",
"sample_len": 1000,
"acc,none": 0.698,
"acc_stderr,none": 0.014526080235459496,
"acc_norm,none": 0.606,
"acc_norm_stderr,none": 0.015459721957493382
},
"arc_easy": {
"name": "arc_easy",
"alias": "arc_easy",
"sample_len": 2376,
"acc,none": 0.4537037037037037,
"acc_stderr,none": 0.010215708295494234,
"acc_norm,none": 0.4099326599326599,
"acc_norm_stderr,none": 0.010091953527506203
},
"winogrande": {
"name": "winogrande",
"alias": "winogrande",
"sample_len": 1267,
"acc,none": 0.5098658247829518,
"acc_stderr,none": 0.014049749833367674
},
"blimp_adjunct_island": {
"name": "blimp_adjunct_island",
"alias": "blimp_adjunct_island",
"sample_len": 1000,
"acc,none": 0.819,
"acc_stderr,none": 0.012181436179178027
},
"blimp_anaphor_gender_agreement": {
"name": "blimp_anaphor_gender_agreement",
"alias": "blimp_anaphor_gender_agreement",
"sample_len": 1000,
"acc,none": 0.841,
"acc_stderr,none": 0.011569479368271195
},
"blimp_anaphor_number_agreement": {
"name": "blimp_anaphor_number_agreement",
"alias": "blimp_anaphor_number_agreement",
"sample_len": 1000,
"acc,none": 0.964,
"acc_stderr,none": 0.00589395781616553
},
"blimp_animate_subject_passive": {
"name": "blimp_animate_subject_passive",
"alias": "blimp_animate_subject_passive",
"sample_len": 1000,
"acc,none": 0.653,
"acc_stderr,none": 0.015060472031706637
},
"blimp_animate_subject_trans": {
"name": "blimp_animate_subject_trans",
"alias": "blimp_animate_subject_trans",
"sample_len": 1000,
"acc,none": 0.915,
"acc_stderr,none": 0.008823426366942316
},
"blimp_causative": {
"name": "blimp_causative",
"alias": "blimp_causative",
"sample_len": 1000,
"acc,none": 0.668,
"acc_stderr,none": 0.014899597242811567
},
"blimp_complex_NP_island": {
"name": "blimp_complex_NP_island",
"alias": "blimp_complex_NP_island",
"sample_len": 1000,
"acc,none": 0.458,
"acc_stderr,none": 0.015763390640483554
},
"blimp_coordinate_structure_constraint_complex_left_branch": {
"name": "blimp_coordinate_structure_constraint_complex_left_branch",
"alias": "blimp_coordinate_structure_constraint_complex_left_branch",
"sample_len": 1000,
"acc,none": 0.344,
"acc_stderr,none": 0.015029633724408919
},
"blimp_coordinate_structure_constraint_object_extraction": {
"name": "blimp_coordinate_structure_constraint_object_extraction",
"alias": "blimp_coordinate_structure_constraint_object_extraction",
"sample_len": 1000,
"acc,none": 0.795,
"acc_stderr,none": 0.012772554096113201
},
"blimp_determiner_noun_agreement_1": {
"name": "blimp_determiner_noun_agreement_1",
"alias": "blimp_determiner_noun_agreement_1",
"sample_len": 1000,
"acc,none": 0.963,
"acc_stderr,none": 0.0059721576223896395
},
"blimp_determiner_noun_agreement_2": {
"name": "blimp_determiner_noun_agreement_2",
"alias": "blimp_determiner_noun_agreement_2",
"sample_len": 1000,
"acc,none": 0.941,
"acc_stderr,none": 0.007454835650406693
},
"blimp_determiner_noun_agreement_irregular_1": {
"name": "blimp_determiner_noun_agreement_irregular_1",
"alias": "blimp_determiner_noun_agreement_irregular_1",
"sample_len": 1000,
"acc,none": 0.85,
"acc_stderr,none": 0.011297239823409418
},
"blimp_determiner_noun_agreement_irregular_2": {
"name": "blimp_determiner_noun_agreement_irregular_2",
"alias": "blimp_determiner_noun_agreement_irregular_2",
"sample_len": 1000,
"acc,none": 0.919,
"acc_stderr,none": 0.008632121032139986
},
"blimp_determiner_noun_agreement_with_adj_2": {
"name": "blimp_determiner_noun_agreement_with_adj_2",
"alias": "blimp_determiner_noun_agreement_with_adj_2",
"sample_len": 1000,
"acc,none": 0.905,
"acc_stderr,none": 0.009276910103103353
},
"blimp_determiner_noun_agreement_with_adj_irregular_1": {
"name": "blimp_determiner_noun_agreement_with_adj_irregular_1",
"alias": "blimp_determiner_noun_agreement_with_adj_irregular_1",
"sample_len": 1000,
"acc,none": 0.805,
"acc_stderr,none": 0.012535235623319256
},
"blimp_determiner_noun_agreement_with_adj_irregular_2": {
"name": "blimp_determiner_noun_agreement_with_adj_irregular_2",
"alias": "blimp_determiner_noun_agreement_with_adj_irregular_2",
"sample_len": 1000,
"acc,none": 0.885,
"acc_stderr,none": 0.010093407594904551
},
"blimp_determiner_noun_agreement_with_adjective_1": {
"name": "blimp_determiner_noun_agreement_with_adjective_1",
"alias": "blimp_determiner_noun_agreement_with_adjective_1",
"sample_len": 1000,
"acc,none": 0.93,
"acc_stderr,none": 0.008072494358323525
},
"blimp_distractor_agreement_relational_noun": {
"name": "blimp_distractor_agreement_relational_noun",
"alias": "blimp_distractor_agreement_relational_noun",
"sample_len": 1000,
"acc,none": 0.809,
"acc_stderr,none": 0.012436787112179482
},
"blimp_distractor_agreement_relative_clause": {
"name": "blimp_distractor_agreement_relative_clause",
"alias": "blimp_distractor_agreement_relative_clause",
"sample_len": 1000,
"acc,none": 0.686,
"acc_stderr,none": 0.014683991951088084
},
"blimp_drop_argument": {
"name": "blimp_drop_argument",
"alias": "blimp_drop_argument",
"sample_len": 1000,
"acc,none": 0.738,
"acc_stderr,none": 0.013912208651021217
},
"blimp_ellipsis_n_bar_1": {
"name": "blimp_ellipsis_n_bar_1",
"alias": "blimp_ellipsis_n_bar_1",
"sample_len": 1000,
"acc,none": 0.813,
"acc_stderr,none": 0.012336254828074168
},
"blimp_ellipsis_n_bar_2": {
"name": "blimp_ellipsis_n_bar_2",
"alias": "blimp_ellipsis_n_bar_2",
"sample_len": 1000,
"acc,none": 0.826,
"acc_stderr,none": 0.011994493230973447
},
"blimp_existential_there_object_raising": {
"name": "blimp_existential_there_object_raising",
"alias": "blimp_existential_there_object_raising",
"sample_len": 1000,
"acc,none": 0.763,
"acc_stderr,none": 0.013454070462577935
},
"blimp_existential_there_quantifiers_1": {
"name": "blimp_existential_there_quantifiers_1",
"alias": "blimp_existential_there_quantifiers_1",
"sample_len": 1000,
"acc,none": 0.957,
"acc_stderr,none": 0.006418114379799739
},
"blimp_existential_there_quantifiers_2": {
"name": "blimp_existential_there_quantifiers_2",
"alias": "blimp_existential_there_quantifiers_2",
"sample_len": 1000,
"acc,none": 0.229,
"acc_stderr,none": 0.013294199326613684
},
"blimp_existential_there_subject_raising": {
"name": "blimp_existential_there_subject_raising",
"alias": "blimp_existential_there_subject_raising",
"sample_len": 1000,
"acc,none": 0.81,
"acc_stderr,none": 0.012411851354816254
},
"blimp_expletive_it_object_raising": {
"name": "blimp_expletive_it_object_raising",
"alias": "blimp_expletive_it_object_raising",
"sample_len": 1000,
"acc,none": 0.761,
"acc_stderr,none": 0.013493000446937704
},
"blimp_inchoative": {
"name": "blimp_inchoative",
"alias": "blimp_inchoative",
"sample_len": 1000,
"acc,none": 0.611,
"acc_stderr,none": 0.015424555647308514
},
"blimp_intransitive": {
"name": "blimp_intransitive",
"alias": "blimp_intransitive",
"sample_len": 1000,
"acc,none": 0.693,
"acc_stderr,none": 0.01459328489285274
},
"blimp_irregular_past_participle_adjectives": {
"name": "blimp_irregular_past_participle_adjectives",
"alias": "blimp_irregular_past_participle_adjectives",
"sample_len": 1000,
"acc,none": 0.908,
"acc_stderr,none": 0.009144376393151129
},
"blimp_irregular_past_participle_verbs": {
"name": "blimp_irregular_past_participle_verbs",
"alias": "blimp_irregular_past_participle_verbs",
"sample_len": 1000,
"acc,none": 0.886,
"acc_stderr,none": 0.010055103435823278
},
"blimp_irregular_plural_subject_verb_agreement_1": {
"name": "blimp_irregular_plural_subject_verb_agreement_1",
"alias": "blimp_irregular_plural_subject_verb_agreement_1",
"sample_len": 1000,
"acc,none": 0.845,
"acc_stderr,none": 0.011450157470799522
},
"blimp_irregular_plural_subject_verb_agreement_2": {
"name": "blimp_irregular_plural_subject_verb_agreement_2",
"alias": "blimp_irregular_plural_subject_verb_agreement_2",
"sample_len": 1000,
"acc,none": 0.881,
"acc_stderr,none": 0.01024421514533661
},
"blimp_left_branch_island_echo_question": {
"name": "blimp_left_branch_island_echo_question",
"alias": "blimp_left_branch_island_echo_question",
"sample_len": 1000,
"acc,none": 0.551,
"acc_stderr,none": 0.015736792768752044
},
"blimp_left_branch_island_simple_question": {
"name": "blimp_left_branch_island_simple_question",
"alias": "blimp_left_branch_island_simple_question",
"sample_len": 1000,
"acc,none": 0.438,
"acc_stderr,none": 0.01569721001969466
},
"blimp_matrix_question_npi_licensor_present": {
"name": "blimp_matrix_question_npi_licensor_present",
"alias": "blimp_matrix_question_npi_licensor_present",
"sample_len": 1000,
"acc,none": 0.571,
"acc_stderr,none": 0.015658997547870177
},
"blimp_npi_present_1": {
"name": "blimp_npi_present_1",
"alias": "blimp_npi_present_1",
"sample_len": 1000,
"acc,none": 0.491,
"acc_stderr,none": 0.0158167369950053
},
"blimp_npi_present_2": {
"name": "blimp_npi_present_2",
"alias": "blimp_npi_present_2",
"sample_len": 1000,
"acc,none": 0.644,
"acc_stderr,none": 0.015149042659306678
},
"blimp_only_npi_licensor_present": {
"name": "blimp_only_npi_licensor_present",
"alias": "blimp_only_npi_licensor_present",
"sample_len": 1000,
"acc,none": 0.569,
"acc_stderr,none": 0.015667944488173442
},
"blimp_only_npi_scope": {
"name": "blimp_only_npi_scope",
"alias": "blimp_only_npi_scope",
"sample_len": 1000,
"acc,none": 0.512,
"acc_stderr,none": 0.015814743314581686
},
"blimp_passive_1": {
"name": "blimp_passive_1",
"alias": "blimp_passive_1",
"sample_len": 1000,
"acc,none": 0.851,
"acc_stderr,none": 0.011266140684632246
},
"blimp_passive_2": {
"name": "blimp_passive_2",
"alias": "blimp_passive_2",
"sample_len": 1000,
"acc,none": 0.851,
"acc_stderr,none": 0.011266140684632246
},
"blimp_principle_A_c_command": {
"name": "blimp_principle_A_c_command",
"alias": "blimp_principle_A_c_command",
"sample_len": 1000,
"acc,none": 0.699,
"acc_stderr,none": 0.01451239503354321
},
"blimp_principle_A_case_1": {
"name": "blimp_principle_A_case_1",
"alias": "blimp_principle_A_case_1",
"sample_len": 1000,
"acc,none": 1.0,
"acc_stderr,none": 0.0
},
"blimp_principle_A_case_2": {
"name": "blimp_principle_A_case_2",
"alias": "blimp_principle_A_case_2",
"sample_len": 1000,
"acc,none": 0.896,
"acc_stderr,none": 0.009658016218524242
},
"blimp_principle_A_domain_1": {
"name": "blimp_principle_A_domain_1",
"alias": "blimp_principle_A_domain_1",
"sample_len": 1000,
"acc,none": 0.848,
"acc_stderr,none": 0.01135891830347525
},
"blimp_principle_A_domain_2": {
"name": "blimp_principle_A_domain_2",
"alias": "blimp_principle_A_domain_2",
"sample_len": 1000,
"acc,none": 0.718,
"acc_stderr,none": 0.01423652621529138
},
"blimp_principle_A_domain_3": {
"name": "blimp_principle_A_domain_3",
"alias": "blimp_principle_A_domain_3",
"sample_len": 1000,
"acc,none": 0.604,
"acc_stderr,none": 0.015473313265859495
},
"blimp_principle_A_reconstruction": {
"name": "blimp_principle_A_reconstruction",
"alias": "blimp_principle_A_reconstruction",
"sample_len": 1000,
"acc,none": 0.321,
"acc_stderr,none": 0.01477082181793475
},
"blimp_regular_plural_subject_verb_agreement_1": {
"name": "blimp_regular_plural_subject_verb_agreement_1",
"alias": "blimp_regular_plural_subject_verb_agreement_1",
"sample_len": 1000,
"acc,none": 0.86,
"acc_stderr,none": 0.010978183844357888
},
"blimp_regular_plural_subject_verb_agreement_2": {
"name": "blimp_regular_plural_subject_verb_agreement_2",
"alias": "blimp_regular_plural_subject_verb_agreement_2",
"sample_len": 1000,
"acc,none": 0.894,
"acc_stderr,none": 0.00973955126578524
},
"blimp_sentential_negation_npi_licensor_present": {
"name": "blimp_sentential_negation_npi_licensor_present",
"alias": "blimp_sentential_negation_npi_licensor_present",
"sample_len": 1000,
"acc,none": 0.998,
"acc_stderr,none": 0.0014135055705578234
},
"blimp_sentential_negation_npi_scope": {
"name": "blimp_sentential_negation_npi_scope",
"alias": "blimp_sentential_negation_npi_scope",
"sample_len": 1000,
"acc,none": 0.6,
"acc_stderr,none": 0.015499685165842571
},
"blimp_sentential_subject_island": {
"name": "blimp_sentential_subject_island",
"alias": "blimp_sentential_subject_island",
"sample_len": 1000,
"acc,none": 0.499,
"acc_stderr,none": 0.015819268290576817
},
"blimp_superlative_quantifiers_1": {
"name": "blimp_superlative_quantifiers_1",
"alias": "blimp_superlative_quantifiers_1",
"sample_len": 1000,
"acc,none": 0.535,
"acc_stderr,none": 0.015780495050030086
},
"blimp_superlative_quantifiers_2": {
"name": "blimp_superlative_quantifiers_2",
"alias": "blimp_superlative_quantifiers_2",
"sample_len": 1000,
"acc,none": 0.471,
"acc_stderr,none": 0.015792669451628764
},
"blimp_tough_vs_raising_1": {
"name": "blimp_tough_vs_raising_1",
"alias": "blimp_tough_vs_raising_1",
"sample_len": 1000,
"acc,none": 0.469,
"acc_stderr,none": 0.015788865959538965
},
"blimp_tough_vs_raising_2": {
"name": "blimp_tough_vs_raising_2",
"alias": "blimp_tough_vs_raising_2",
"sample_len": 1000,
"acc,none": 0.854,
"acc_stderr,none": 0.011171786285496551
},
"blimp_transitive": {
"name": "blimp_transitive",
"alias": "blimp_transitive",
"sample_len": 1000,
"acc,none": 0.783,
"acc_stderr,none": 0.013041513757270706
},
"blimp_wh_island": {
"name": "blimp_wh_island",
"alias": "blimp_wh_island",
"sample_len": 1000,
"acc,none": 0.743,
"acc_stderr,none": 0.013825416526895026
},
"blimp_wh_questions_object_gap": {
"name": "blimp_wh_questions_object_gap",
"alias": "blimp_wh_questions_object_gap",
"sample_len": 1000,
"acc,none": 0.708,
"acc_stderr,none": 0.014385511563477433
},
"blimp_wh_questions_subject_gap": {
"name": "blimp_wh_questions_subject_gap",
"alias": "blimp_wh_questions_subject_gap",
"sample_len": 1000,
"acc,none": 0.911,
"acc_stderr,none": 0.009008893392651516
},
"blimp_wh_questions_subject_gap_long_distance": {
"name": "blimp_wh_questions_subject_gap_long_distance",
"alias": "blimp_wh_questions_subject_gap_long_distance",
"sample_len": 1000,
"acc,none": 0.807,
"acc_stderr,none": 0.012486268734370044
},
"blimp_wh_vs_that_no_gap": {
"name": "blimp_wh_vs_that_no_gap",
"alias": "blimp_wh_vs_that_no_gap",
"sample_len": 1000,
"acc,none": 0.966,
"acc_stderr,none": 0.0057338361396953854
},
"blimp_wh_vs_that_no_gap_long_distance": {
"name": "blimp_wh_vs_that_no_gap_long_distance",
"alias": "blimp_wh_vs_that_no_gap_long_distance",
"sample_len": 1000,
"acc,none": 0.958,
"acc_stderr,none": 0.006346359293033784
},
"blimp_wh_vs_that_with_gap": {
"name": "blimp_wh_vs_that_with_gap",
"alias": "blimp_wh_vs_that_with_gap",
"sample_len": 1000,
"acc,none": 0.38,
"acc_stderr,none": 0.015356947477797658
},
"blimp_wh_vs_that_with_gap_long_distance": {
"name": "blimp_wh_vs_that_with_gap_long_distance",
"alias": "blimp_wh_vs_that_with_gap_long_distance",
"sample_len": 1000,
"acc,none": 0.2,
"acc_stderr,none": 0.012655439943366655
},
"blimp": {
"alias": "blimp",
"name": "blimp",
"sample_len": 67000,
"acc,none": 0.7324029850746269,
"acc_stderr,none": 0.0015343189725241936,
"sample_count": {
"acc,none": 67000
}
}
}
}