Codemind-study / m23 /benchmark_results.json
B4K2xx's picture
Add m23 weights and results
1c2639a verified
Raw
History Blame Contribute Delete
18.5 kB
{
"model_id": "m23",
"checkpoint": "results/m23/final.pt",
"trained_step": 15258,
"num_fewshot": 0,
"limit": null,
"results": {
"lambada_openai": {
"name": "lambada_openai",
"alias": "lambada_openai",
"sample_len": 5153,
"perplexity,none": 657.6825390310986,
"perplexity_stderr,none": 36.041339126600164,
"acc,none": 0.16922181253638657,
"acc_stderr,none": 0.005223754842464426
},
"hellaswag": {
"name": "hellaswag",
"alias": "hellaswag",
"sample_len": 10042,
"acc,none": 0.2672774347739494,
"acc_stderr,none": 0.004416339450436367,
"acc_norm,none": 0.2771360286795459,
"acc_norm_stderr,none": 0.004466695023677567
},
"piqa": {
"name": "piqa",
"alias": "piqa",
"sample_len": 1838,
"acc,none": 0.5952121871599565,
"acc_stderr,none": 0.011452361375057018,
"acc_norm,none": 0.5892274211099021,
"acc_norm_stderr,none": 0.011478565556775618
},
"sciq": {
"name": "sciq",
"alias": "sciq",
"sample_len": 1000,
"acc,none": 0.68,
"acc_stderr,none": 0.014758652303574938,
"acc_norm,none": 0.591,
"acc_norm_stderr,none": 0.015555094373257946
},
"arc_easy": {
"name": "arc_easy",
"alias": "arc_easy",
"sample_len": 2376,
"acc,none": 0.4494949494949495,
"acc_stderr,none": 0.010207308833916014,
"acc_norm,none": 0.406986531986532,
"acc_norm_stderr,none": 0.010080695355466778
},
"winogrande": {
"name": "winogrande",
"alias": "winogrande",
"sample_len": 1267,
"acc,none": 0.5074980268350434,
"acc_stderr,none": 0.014050905521228646
},
"blimp_adjunct_island": {
"name": "blimp_adjunct_island",
"alias": "blimp_adjunct_island",
"sample_len": 1000,
"acc,none": 0.787,
"acc_stderr,none": 0.012953717566737312
},
"blimp_anaphor_gender_agreement": {
"name": "blimp_anaphor_gender_agreement",
"alias": "blimp_anaphor_gender_agreement",
"sample_len": 1000,
"acc,none": 0.958,
"acc_stderr,none": 0.006346359293033784
},
"blimp_anaphor_number_agreement": {
"name": "blimp_anaphor_number_agreement",
"alias": "blimp_anaphor_number_agreement",
"sample_len": 1000,
"acc,none": 0.982,
"acc_stderr,none": 0.00420638724961144
},
"blimp_animate_subject_passive": {
"name": "blimp_animate_subject_passive",
"alias": "blimp_animate_subject_passive",
"sample_len": 1000,
"acc,none": 0.741,
"acc_stderr,none": 0.013860415257527861
},
"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": 0.704,
"acc_stderr,none": 0.01444273494157504
},
"blimp_complex_NP_island": {
"name": "blimp_complex_NP_island",
"alias": "blimp_complex_NP_island",
"sample_len": 1000,
"acc,none": 0.555,
"acc_stderr,none": 0.015723301886761007
},
"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": 0.447,
"acc_stderr,none": 0.015730176046009098
},
"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": 0.75,
"acc_stderr,none": 0.013699915608779773
},
"blimp_determiner_noun_agreement_1": {
"name": "blimp_determiner_noun_agreement_1",
"alias": "blimp_determiner_noun_agreement_1",
"sample_len": 1000,
"acc,none": 0.971,
"acc_stderr,none": 0.005309160685756948
},
"blimp_determiner_noun_agreement_2": {
"name": "blimp_determiner_noun_agreement_2",
"alias": "blimp_determiner_noun_agreement_2",
"sample_len": 1000,
"acc,none": 0.949,
"acc_stderr,none": 0.0069604200625714725
},
"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": 0.857,
"acc_stderr,none": 0.011075814808567074
},
"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": 0.937,
"acc_stderr,none": 0.007687007876286461
},
"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": 0.915,
"acc_stderr,none": 0.008823426366942316
},
"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": 0.857,
"acc_stderr,none": 0.011075814808567074
},
"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": 0.908,
"acc_stderr,none": 0.009144376393151129
},
"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": 0.939,
"acc_stderr,none": 0.007572076091557476
},
"blimp_distractor_agreement_relational_noun": {
"name": "blimp_distractor_agreement_relational_noun",
"alias": "blimp_distractor_agreement_relational_noun",
"sample_len": 1000,
"acc,none": 0.844,
"acc_stderr,none": 0.011480235006122295
},
"blimp_distractor_agreement_relative_clause": {
"name": "blimp_distractor_agreement_relative_clause",
"alias": "blimp_distractor_agreement_relative_clause",
"sample_len": 1000,
"acc,none": 0.763,
"acc_stderr,none": 0.013454070462577935
},
"blimp_drop_argument": {
"name": "blimp_drop_argument",
"alias": "blimp_drop_argument",
"sample_len": 1000,
"acc,none": 0.751,
"acc_stderr,none": 0.013681600278702275
},
"blimp_ellipsis_n_bar_1": {
"name": "blimp_ellipsis_n_bar_1",
"alias": "blimp_ellipsis_n_bar_1",
"sample_len": 1000,
"acc,none": 0.772,
"acc_stderr,none": 0.013273740700804434
},
"blimp_ellipsis_n_bar_2": {
"name": "blimp_ellipsis_n_bar_2",
"alias": "blimp_ellipsis_n_bar_2",
"sample_len": 1000,
"acc,none": 0.843,
"acc_stderr,none": 0.011510146979230262
},
"blimp_existential_there_object_raising": {
"name": "blimp_existential_there_object_raising",
"alias": "blimp_existential_there_object_raising",
"sample_len": 1000,
"acc,none": 0.821,
"acc_stderr,none": 0.012128730605719056
},
"blimp_existential_there_quantifiers_1": {
"name": "blimp_existential_there_quantifiers_1",
"alias": "blimp_existential_there_quantifiers_1",
"sample_len": 1000,
"acc,none": 0.978,
"acc_stderr,none": 0.004640855259274689
},
"blimp_existential_there_quantifiers_2": {
"name": "blimp_existential_there_quantifiers_2",
"alias": "blimp_existential_there_quantifiers_2",
"sample_len": 1000,
"acc,none": 0.114,
"acc_stderr,none": 0.010055103435823278
},
"blimp_existential_there_subject_raising": {
"name": "blimp_existential_there_subject_raising",
"alias": "blimp_existential_there_subject_raising",
"sample_len": 1000,
"acc,none": 0.814,
"acc_stderr,none": 0.012310790208412926
},
"blimp_expletive_it_object_raising": {
"name": "blimp_expletive_it_object_raising",
"alias": "blimp_expletive_it_object_raising",
"sample_len": 1000,
"acc,none": 0.768,
"acc_stderr,none": 0.013354937452281623
},
"blimp_inchoative": {
"name": "blimp_inchoative",
"alias": "blimp_inchoative",
"sample_len": 1000,
"acc,none": 0.641,
"acc_stderr,none": 0.01517726422479853
},
"blimp_intransitive": {
"name": "blimp_intransitive",
"alias": "blimp_intransitive",
"sample_len": 1000,
"acc,none": 0.745,
"acc_stderr,none": 0.013790038620872863
},
"blimp_irregular_past_participle_adjectives": {
"name": "blimp_irregular_past_participle_adjectives",
"alias": "blimp_irregular_past_participle_adjectives",
"sample_len": 1000,
"acc,none": 0.751,
"acc_stderr,none": 0.013681600278702275
},
"blimp_irregular_past_participle_verbs": {
"name": "blimp_irregular_past_participle_verbs",
"alias": "blimp_irregular_past_participle_verbs",
"sample_len": 1000,
"acc,none": 0.862,
"acc_stderr,none": 0.010912152632504508
},
"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": 0.877,
"acc_stderr,none": 0.010391293421849803
},
"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": 0.861,
"acc_stderr,none": 0.010945263761042892
},
"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": 0.505,
"acc_stderr,none": 0.015818508944436743
},
"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": 0.449,
"acc_stderr,none": 0.015736792768752044
},
"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": 0.46,
"acc_stderr,none": 0.015768596914394403
},
"blimp_npi_present_1": {
"name": "blimp_npi_present_1",
"alias": "blimp_npi_present_1",
"sample_len": 1000,
"acc,none": 0.603,
"acc_stderr,none": 0.015480007449307982
},
"blimp_npi_present_2": {
"name": "blimp_npi_present_2",
"alias": "blimp_npi_present_2",
"sample_len": 1000,
"acc,none": 0.595,
"acc_stderr,none": 0.01553113699045296
},
"blimp_only_npi_licensor_present": {
"name": "blimp_only_npi_licensor_present",
"alias": "blimp_only_npi_licensor_present",
"sample_len": 1000,
"acc,none": 0.905,
"acc_stderr,none": 0.009276910103103353
},
"blimp_only_npi_scope": {
"name": "blimp_only_npi_scope",
"alias": "blimp_only_npi_scope",
"sample_len": 1000,
"acc,none": 0.673,
"acc_stderr,none": 0.014842213153411162
},
"blimp_passive_1": {
"name": "blimp_passive_1",
"alias": "blimp_passive_1",
"sample_len": 1000,
"acc,none": 0.879,
"acc_stderr,none": 0.010318210380946179
},
"blimp_passive_2": {
"name": "blimp_passive_2",
"alias": "blimp_passive_2",
"sample_len": 1000,
"acc,none": 0.869,
"acc_stderr,none": 0.0106748748448379
},
"blimp_principle_A_c_command": {
"name": "blimp_principle_A_c_command",
"alias": "blimp_principle_A_c_command",
"sample_len": 1000,
"acc,none": 0.628,
"acc_stderr,none": 0.01529214994204052
},
"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": 0.884,
"acc_stderr,none": 0.010131468138756922
},
"blimp_principle_A_domain_1": {
"name": "blimp_principle_A_domain_1",
"alias": "blimp_principle_A_domain_1",
"sample_len": 1000,
"acc,none": 0.951,
"acc_stderr,none": 0.006829761756140959
},
"blimp_principle_A_domain_2": {
"name": "blimp_principle_A_domain_2",
"alias": "blimp_principle_A_domain_2",
"sample_len": 1000,
"acc,none": 0.729,
"acc_stderr,none": 0.014062601350986125
},
"blimp_principle_A_domain_3": {
"name": "blimp_principle_A_domain_3",
"alias": "blimp_principle_A_domain_3",
"sample_len": 1000,
"acc,none": 0.625,
"acc_stderr,none": 0.015316971293620996
},
"blimp_principle_A_reconstruction": {
"name": "blimp_principle_A_reconstruction",
"alias": "blimp_principle_A_reconstruction",
"sample_len": 1000,
"acc,none": 0.467,
"acc_stderr,none": 0.015784807891138876
},
"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": 0.848,
"acc_stderr,none": 0.01135891830347525
},
"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": 0.871,
"acc_stderr,none": 0.010605256784796466
},
"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": 0.995,
"acc_stderr,none": 0.0022315868748448895
},
"blimp_sentential_negation_npi_scope": {
"name": "blimp_sentential_negation_npi_scope",
"alias": "blimp_sentential_negation_npi_scope",
"sample_len": 1000,
"acc,none": 0.652,
"acc_stderr,none": 0.015070604603768328
},
"blimp_sentential_subject_island": {
"name": "blimp_sentential_subject_island",
"alias": "blimp_sentential_subject_island",
"sample_len": 1000,
"acc,none": 0.493,
"acc_stderr,none": 0.01581774956184353
},
"blimp_superlative_quantifiers_1": {
"name": "blimp_superlative_quantifiers_1",
"alias": "blimp_superlative_quantifiers_1",
"sample_len": 1000,
"acc,none": 0.822,
"acc_stderr,none": 0.012102167676183571
},
"blimp_superlative_quantifiers_2": {
"name": "blimp_superlative_quantifiers_2",
"alias": "blimp_superlative_quantifiers_2",
"sample_len": 1000,
"acc,none": 0.731,
"acc_stderr,none": 0.014029819522568203
},
"blimp_tough_vs_raising_1": {
"name": "blimp_tough_vs_raising_1",
"alias": "blimp_tough_vs_raising_1",
"sample_len": 1000,
"acc,none": 0.543,
"acc_stderr,none": 0.0157606915901365
},
"blimp_tough_vs_raising_2": {
"name": "blimp_tough_vs_raising_2",
"alias": "blimp_tough_vs_raising_2",
"sample_len": 1000,
"acc,none": 0.898,
"acc_stderr,none": 0.009575368801653944
},
"blimp_transitive": {
"name": "blimp_transitive",
"alias": "blimp_transitive",
"sample_len": 1000,
"acc,none": 0.812,
"acc_stderr,none": 0.012361586015103758
},
"blimp_wh_island": {
"name": "blimp_wh_island",
"alias": "blimp_wh_island",
"sample_len": 1000,
"acc,none": 0.719,
"acc_stderr,none": 0.01422115470843494
},
"blimp_wh_questions_object_gap": {
"name": "blimp_wh_questions_object_gap",
"alias": "blimp_wh_questions_object_gap",
"sample_len": 1000,
"acc,none": 0.684,
"acc_stderr,none": 0.014709193056057166
},
"blimp_wh_questions_subject_gap": {
"name": "blimp_wh_questions_subject_gap",
"alias": "blimp_wh_questions_subject_gap",
"sample_len": 1000,
"acc,none": 0.877,
"acc_stderr,none": 0.010391293421849803
},
"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": 0.849,
"acc_stderr,none": 0.011328165223341657
},
"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": 0.948,
"acc_stderr,none": 0.007024624213817088
},
"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": 0.966,
"acc_stderr,none": 0.0057338361396953854
},
"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": 0.526,
"acc_stderr,none": 0.015797897758042797
},
"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": 0.23,
"acc_stderr,none": 0.01331455133593608
},
"blimp": {
"alias": "blimp",
"name": "blimp",
"sample_len": 67000,
"acc,none": 0.761955223880597,
"acc_stderr,none": 0.0014852092525077722,
"sample_count": {
"acc,none": 67000
}
}
}
}