Codemind-study / m26 /benchmark_results.json
B4K2xx's picture
Add m26 weights and results
2f5738c verified
Raw
History Blame Contribute Delete
18.5 kB
{
"model_id": "m26",
"checkpoint": "results/m26/final.pt",
"trained_step": 15258,
"num_fewshot": 0,
"limit": null,
"results": {
"lambada_openai": {
"name": "lambada_openai",
"alias": "lambada_openai",
"sample_len": 5153,
"perplexity,none": 595.4815310921231,
"perplexity_stderr,none": 32.89384889135734,
"acc,none": 0.17950708325247428,
"acc_stderr,none": 0.0053467550223617434
},
"hellaswag": {
"name": "hellaswag",
"alias": "hellaswag",
"sample_len": 10042,
"acc,none": 0.27165903206532566,
"acc_stderr,none": 0.004439059440526545,
"acc_norm,none": 0.27823142800239,
"acc_norm_stderr,none": 0.004472121485161612
},
"piqa": {
"name": "piqa",
"alias": "piqa",
"sample_len": 1838,
"acc,none": 0.5979325353645266,
"acc_stderr,none": 0.011439867127267474,
"acc_norm,none": 0.5816104461371056,
"acc_norm_stderr,none": 0.011509380214588163
},
"sciq": {
"name": "sciq",
"alias": "sciq",
"sample_len": 1000,
"acc,none": 0.663,
"acc_stderr,none": 0.01495508791865346,
"acc_norm,none": 0.579,
"acc_norm_stderr,none": 0.015620595475301287
},
"arc_easy": {
"name": "arc_easy",
"alias": "arc_easy",
"sample_len": 2376,
"acc,none": 0.4511784511784512,
"acc_stderr,none": 0.010210757101073465,
"acc_norm,none": 0.4010942760942761,
"acc_norm_stderr,none": 0.010057051106534466
},
"winogrande": {
"name": "winogrande",
"alias": "winogrande",
"sample_len": 1267,
"acc,none": 0.5303867403314917,
"acc_stderr,none": 0.014026510839428848
},
"blimp_adjunct_island": {
"name": "blimp_adjunct_island",
"alias": "blimp_adjunct_island",
"sample_len": 1000,
"acc,none": 0.773,
"acc_stderr,none": 0.013253174964763977
},
"blimp_anaphor_gender_agreement": {
"name": "blimp_anaphor_gender_agreement",
"alias": "blimp_anaphor_gender_agreement",
"sample_len": 1000,
"acc,none": 0.931,
"acc_stderr,none": 0.008018934050315124
},
"blimp_anaphor_number_agreement": {
"name": "blimp_anaphor_number_agreement",
"alias": "blimp_anaphor_number_agreement",
"sample_len": 1000,
"acc,none": 0.985,
"acc_stderr,none": 0.003845749574502972
},
"blimp_animate_subject_passive": {
"name": "blimp_animate_subject_passive",
"alias": "blimp_animate_subject_passive",
"sample_len": 1000,
"acc,none": 0.711,
"acc_stderr,none": 0.014341711358296287
},
"blimp_animate_subject_trans": {
"name": "blimp_animate_subject_trans",
"alias": "blimp_animate_subject_trans",
"sample_len": 1000,
"acc,none": 0.903,
"acc_stderr,none": 0.009363689373248062
},
"blimp_causative": {
"name": "blimp_causative",
"alias": "blimp_causative",
"sample_len": 1000,
"acc,none": 0.705,
"acc_stderr,none": 0.014428554438445425
},
"blimp_complex_NP_island": {
"name": "blimp_complex_NP_island",
"alias": "blimp_complex_NP_island",
"sample_len": 1000,
"acc,none": 0.437,
"acc_stderr,none": 0.015693223928730425
},
"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": 0.468,
"acc_stderr,none": 0.015786868759359002
},
"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": 0.738,
"acc_stderr,none": 0.013912208651021217
},
"blimp_determiner_noun_agreement_1": {
"name": "blimp_determiner_noun_agreement_1",
"alias": "blimp_determiner_noun_agreement_1",
"sample_len": 1000,
"acc,none": 0.962,
"acc_stderr,none": 0.0060491811505849775
},
"blimp_determiner_noun_agreement_2": {
"name": "blimp_determiner_noun_agreement_2",
"alias": "blimp_determiner_noun_agreement_2",
"sample_len": 1000,
"acc,none": 0.95,
"acc_stderr,none": 0.006895472974897918
},
"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": 0.898,
"acc_stderr,none": 0.009575368801653944
},
"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": 0.893,
"acc_stderr,none": 0.009779910359847203
},
"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": 0.91,
"acc_stderr,none": 0.009054390204866477
},
"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": 0.839,
"acc_stderr,none": 0.011628164696727263
},
"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": 0.886,
"acc_stderr,none": 0.010055103435823278
},
"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": 0.912,
"acc_stderr,none": 0.008963053962592012
},
"blimp_distractor_agreement_relational_noun": {
"name": "blimp_distractor_agreement_relational_noun",
"alias": "blimp_distractor_agreement_relational_noun",
"sample_len": 1000,
"acc,none": 0.847,
"acc_stderr,none": 0.011389500459665504
},
"blimp_distractor_agreement_relative_clause": {
"name": "blimp_distractor_agreement_relative_clause",
"alias": "blimp_distractor_agreement_relative_clause",
"sample_len": 1000,
"acc,none": 0.722,
"acc_stderr,none": 0.014174516461485272
},
"blimp_drop_argument": {
"name": "blimp_drop_argument",
"alias": "blimp_drop_argument",
"sample_len": 1000,
"acc,none": 0.766,
"acc_stderr,none": 0.01339490288966006
},
"blimp_ellipsis_n_bar_1": {
"name": "blimp_ellipsis_n_bar_1",
"alias": "blimp_ellipsis_n_bar_1",
"sample_len": 1000,
"acc,none": 0.752,
"acc_stderr,none": 0.013663187134877524
},
"blimp_ellipsis_n_bar_2": {
"name": "blimp_ellipsis_n_bar_2",
"alias": "blimp_ellipsis_n_bar_2",
"sample_len": 1000,
"acc,none": 0.879,
"acc_stderr,none": 0.010318210380946179
},
"blimp_existential_there_object_raising": {
"name": "blimp_existential_there_object_raising",
"alias": "blimp_existential_there_object_raising",
"sample_len": 1000,
"acc,none": 0.779,
"acc_stderr,none": 0.01312750285969618
},
"blimp_existential_there_quantifiers_1": {
"name": "blimp_existential_there_quantifiers_1",
"alias": "blimp_existential_there_quantifiers_1",
"sample_len": 1000,
"acc,none": 0.97,
"acc_stderr,none": 0.0053971408290991564
},
"blimp_existential_there_quantifiers_2": {
"name": "blimp_existential_there_quantifiers_2",
"alias": "blimp_existential_there_quantifiers_2",
"sample_len": 1000,
"acc,none": 0.35,
"acc_stderr,none": 0.015090650341444127
},
"blimp_existential_there_subject_raising": {
"name": "blimp_existential_there_subject_raising",
"alias": "blimp_existential_there_subject_raising",
"sample_len": 1000,
"acc,none": 0.892,
"acc_stderr,none": 0.009820001651345708
},
"blimp_expletive_it_object_raising": {
"name": "blimp_expletive_it_object_raising",
"alias": "blimp_expletive_it_object_raising",
"sample_len": 1000,
"acc,none": 0.766,
"acc_stderr,none": 0.01339490288966006
},
"blimp_inchoative": {
"name": "blimp_inchoative",
"alias": "blimp_inchoative",
"sample_len": 1000,
"acc,none": 0.673,
"acc_stderr,none": 0.014842213153411162
},
"blimp_intransitive": {
"name": "blimp_intransitive",
"alias": "blimp_intransitive",
"sample_len": 1000,
"acc,none": 0.742,
"acc_stderr,none": 0.013842963108656601
},
"blimp_irregular_past_participle_adjectives": {
"name": "blimp_irregular_past_participle_adjectives",
"alias": "blimp_irregular_past_participle_adjectives",
"sample_len": 1000,
"acc,none": 0.911,
"acc_stderr,none": 0.009008893392651516
},
"blimp_irregular_past_participle_verbs": {
"name": "blimp_irregular_past_participle_verbs",
"alias": "blimp_irregular_past_participle_verbs",
"sample_len": 1000,
"acc,none": 0.794,
"acc_stderr,none": 0.012795613612786522
},
"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": 0.875,
"acc_stderr,none": 0.010463483381956722
},
"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": 0.883,
"acc_stderr,none": 0.010169287802713345
},
"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": 0.548,
"acc_stderr,none": 0.01574623586588058
},
"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": 0.536,
"acc_stderr,none": 0.015778243024904673
},
"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": 0.566,
"acc_stderr,none": 0.01568087656637504
},
"blimp_npi_present_1": {
"name": "blimp_npi_present_1",
"alias": "blimp_npi_present_1",
"sample_len": 1000,
"acc,none": 0.54,
"acc_stderr,none": 0.0157685969143944
},
"blimp_npi_present_2": {
"name": "blimp_npi_present_2",
"alias": "blimp_npi_present_2",
"sample_len": 1000,
"acc,none": 0.669,
"acc_stderr,none": 0.0148882725882038
},
"blimp_only_npi_licensor_present": {
"name": "blimp_only_npi_licensor_present",
"alias": "blimp_only_npi_licensor_present",
"sample_len": 1000,
"acc,none": 0.894,
"acc_stderr,none": 0.00973955126578524
},
"blimp_only_npi_scope": {
"name": "blimp_only_npi_scope",
"alias": "blimp_only_npi_scope",
"sample_len": 1000,
"acc,none": 0.638,
"acc_stderr,none": 0.015204840912919383
},
"blimp_passive_1": {
"name": "blimp_passive_1",
"alias": "blimp_passive_1",
"sample_len": 1000,
"acc,none": 0.844,
"acc_stderr,none": 0.011480235006122295
},
"blimp_passive_2": {
"name": "blimp_passive_2",
"alias": "blimp_passive_2",
"sample_len": 1000,
"acc,none": 0.844,
"acc_stderr,none": 0.011480235006122295
},
"blimp_principle_A_c_command": {
"name": "blimp_principle_A_c_command",
"alias": "blimp_principle_A_c_command",
"sample_len": 1000,
"acc,none": 0.58,
"acc_stderr,none": 0.015615500115072966
},
"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": 0.887,
"acc_stderr,none": 0.010016552866696773
},
"blimp_principle_A_domain_1": {
"name": "blimp_principle_A_domain_1",
"alias": "blimp_principle_A_domain_1",
"sample_len": 1000,
"acc,none": 0.95,
"acc_stderr,none": 0.006895472974897918
},
"blimp_principle_A_domain_2": {
"name": "blimp_principle_A_domain_2",
"alias": "blimp_principle_A_domain_2",
"sample_len": 1000,
"acc,none": 0.746,
"acc_stderr,none": 0.013772206565168622
},
"blimp_principle_A_domain_3": {
"name": "blimp_principle_A_domain_3",
"alias": "blimp_principle_A_domain_3",
"sample_len": 1000,
"acc,none": 0.644,
"acc_stderr,none": 0.015149042659306678
},
"blimp_principle_A_reconstruction": {
"name": "blimp_principle_A_reconstruction",
"alias": "blimp_principle_A_reconstruction",
"sample_len": 1000,
"acc,none": 0.649,
"acc_stderr,none": 0.015100563798316508
},
"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": 0.866,
"acc_stderr,none": 0.010777762298369738
},
"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": 0.874,
"acc_stderr,none": 0.010499249222407964
},
"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": 0.697,
"acc_stderr,none": 0.014539683710535193
},
"blimp_sentential_subject_island": {
"name": "blimp_sentential_subject_island",
"alias": "blimp_sentential_subject_island",
"sample_len": 1000,
"acc,none": 0.513,
"acc_stderr,none": 0.015813952101896636
},
"blimp_superlative_quantifiers_1": {
"name": "blimp_superlative_quantifiers_1",
"alias": "blimp_superlative_quantifiers_1",
"sample_len": 1000,
"acc,none": 0.921,
"acc_stderr,none": 0.00853415677333337
},
"blimp_superlative_quantifiers_2": {
"name": "blimp_superlative_quantifiers_2",
"alias": "blimp_superlative_quantifiers_2",
"sample_len": 1000,
"acc,none": 0.691,
"acc_stderr,none": 0.014619600977206347
},
"blimp_tough_vs_raising_1": {
"name": "blimp_tough_vs_raising_1",
"alias": "blimp_tough_vs_raising_1",
"sample_len": 1000,
"acc,none": 0.634,
"acc_stderr,none": 0.0152406127264056
},
"blimp_tough_vs_raising_2": {
"name": "blimp_tough_vs_raising_2",
"alias": "blimp_tough_vs_raising_2",
"sample_len": 1000,
"acc,none": 0.853,
"acc_stderr,none": 0.011203415395160376
},
"blimp_transitive": {
"name": "blimp_transitive",
"alias": "blimp_transitive",
"sample_len": 1000,
"acc,none": 0.825,
"acc_stderr,none": 0.012021627157731998
},
"blimp_wh_island": {
"name": "blimp_wh_island",
"alias": "blimp_wh_island",
"sample_len": 1000,
"acc,none": 0.818,
"acc_stderr,none": 0.0122075806376622
},
"blimp_wh_questions_object_gap": {
"name": "blimp_wh_questions_object_gap",
"alias": "blimp_wh_questions_object_gap",
"sample_len": 1000,
"acc,none": 0.775,
"acc_stderr,none": 0.013211720158614833
},
"blimp_wh_questions_subject_gap": {
"name": "blimp_wh_questions_subject_gap",
"alias": "blimp_wh_questions_subject_gap",
"sample_len": 1000,
"acc,none": 0.941,
"acc_stderr,none": 0.007454835650406693
},
"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": 0.894,
"acc_stderr,none": 0.00973955126578524
},
"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": 0.976,
"acc_stderr,none": 0.004842256441727084
},
"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": 0.982,
"acc_stderr,none": 0.00420638724961144
},
"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": 0.455,
"acc_stderr,none": 0.015755101498347232
},
"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": 0.202,
"acc_stderr,none": 0.01270265158765513
},
"blimp": {
"alias": "blimp",
"name": "blimp",
"sample_len": 67000,
"acc,none": 0.7754029850746268,
"acc_stderr,none": 0.0014728963305261126,
"sample_count": {
"acc,none": 67000
}
}
}
}