{ "results": { "blimp": { "acc,none": 0.7272835820895525, "acc_stderr,none": 0.19867610993066823, "alias": "blimp" }, "blimp_adjunct_island": { "acc,none": 0.753, "acc_stderr,none": 0.013644675781314128, "alias": " - blimp_adjunct_island" }, "blimp_anaphor_gender_agreement": { "acc,none": 0.803, "acc_stderr,none": 0.012583693787968118, "alias": " - blimp_anaphor_gender_agreement" }, "blimp_anaphor_number_agreement": { "acc,none": 0.939, "acc_stderr,none": 0.007572076091557415, "alias": " - blimp_anaphor_number_agreement" }, "blimp_animate_subject_passive": { "acc,none": 0.782, "acc_stderr,none": 0.013063179040595277, "alias": " - blimp_animate_subject_passive" }, "blimp_animate_subject_trans": { "acc,none": 0.816, "acc_stderr,none": 0.012259457340938595, "alias": " - blimp_animate_subject_trans" }, "blimp_causative": { "acc,none": 0.705, "acc_stderr,none": 0.014428554438445514, "alias": " - blimp_causative" }, "blimp_complex_NP_island": { "acc,none": 0.556, "acc_stderr,none": 0.01571976816340209, "alias": " - blimp_complex_NP_island" }, "blimp_coordinate_structure_constraint_complex_left_branch": { "acc,none": 0.487, "acc_stderr,none": 0.01581395210189663, "alias": " - blimp_coordinate_structure_constraint_complex_left_branch" }, "blimp_coordinate_structure_constraint_object_extraction": { "acc,none": 0.695, "acc_stderr,none": 0.014566646394664384, "alias": " - blimp_coordinate_structure_constraint_object_extraction" }, "blimp_determiner_noun_agreement_1": { "acc,none": 0.99, "acc_stderr,none": 0.003148000938676773, "alias": " - blimp_determiner_noun_agreement_1" }, "blimp_determiner_noun_agreement_2": { "acc,none": 0.986, "acc_stderr,none": 0.0037172325482565955, "alias": " - blimp_determiner_noun_agreement_2" }, "blimp_determiner_noun_agreement_irregular_1": { "acc,none": 0.896, "acc_stderr,none": 0.0096580162185243, "alias": " - blimp_determiner_noun_agreement_irregular_1" }, "blimp_determiner_noun_agreement_irregular_2": { "acc,none": 0.942, "acc_stderr,none": 0.007395315455792957, "alias": " - blimp_determiner_noun_agreement_irregular_2" }, "blimp_determiner_noun_agreement_with_adj_2": { "acc,none": 0.946, "acc_stderr,none": 0.0071508835212954446, "alias": " - blimp_determiner_noun_agreement_with_adj_2" }, "blimp_determiner_noun_agreement_with_adj_irregular_1": { "acc,none": 0.852, "acc_stderr,none": 0.011234866364235261, "alias": " - blimp_determiner_noun_agreement_with_adj_irregular_1" }, "blimp_determiner_noun_agreement_with_adj_irregular_2": { "acc,none": 0.902, "acc_stderr,none": 0.009406619184621231, "alias": " - blimp_determiner_noun_agreement_with_adj_irregular_2" }, "blimp_determiner_noun_agreement_with_adjective_1": { "acc,none": 0.953, "acc_stderr,none": 0.006695956678163046, "alias": " - blimp_determiner_noun_agreement_with_adjective_1" }, "blimp_distractor_agreement_relational_noun": { "acc,none": 0.571, "acc_stderr,none": 0.015658997547870247, "alias": " - blimp_distractor_agreement_relational_noun" }, "blimp_distractor_agreement_relative_clause": { "acc,none": 0.51, "acc_stderr,none": 0.015816135752773207, "alias": " - blimp_distractor_agreement_relative_clause" }, "blimp_drop_argument": { "acc,none": 0.78, "acc_stderr,none": 0.013106173040661768, "alias": " - blimp_drop_argument" }, "blimp_ellipsis_n_bar_1": { "acc,none": 0.7, "acc_stderr,none": 0.014498627873361432, "alias": " - blimp_ellipsis_n_bar_1" }, "blimp_ellipsis_n_bar_2": { "acc,none": 0.645, "acc_stderr,none": 0.01513949154378053, "alias": " - blimp_ellipsis_n_bar_2" }, "blimp_existential_there_object_raising": { "acc,none": 0.66, "acc_stderr,none": 0.014987482264363935, "alias": " - blimp_existential_there_object_raising" }, "blimp_existential_there_quantifiers_1": { "acc,none": 0.939, "acc_stderr,none": 0.007572076091557425, "alias": " - blimp_existential_there_quantifiers_1" }, "blimp_existential_there_quantifiers_2": { "acc,none": 0.545, "acc_stderr,none": 0.01575510149834709, "alias": " - blimp_existential_there_quantifiers_2" }, "blimp_existential_there_subject_raising": { "acc,none": 0.836, "acc_stderr,none": 0.011715000693181318, "alias": " - blimp_existential_there_subject_raising" }, "blimp_expletive_it_object_raising": { "acc,none": 0.714, "acc_stderr,none": 0.01429714686251791, "alias": " - blimp_expletive_it_object_raising" }, "blimp_inchoative": { "acc,none": 0.593, "acc_stderr,none": 0.015543249100255545, "alias": " - blimp_inchoative" }, "blimp_intransitive": { "acc,none": 0.752, "acc_stderr,none": 0.013663187134877639, "alias": " - blimp_intransitive" }, "blimp_irregular_past_participle_adjectives": { "acc,none": 0.962, "acc_stderr,none": 0.0060491811505849384, "alias": " - blimp_irregular_past_participle_adjectives" }, "blimp_irregular_past_participle_verbs": { "acc,none": 0.85, "acc_stderr,none": 0.0112972398234093, "alias": " - blimp_irregular_past_participle_verbs" }, "blimp_irregular_plural_subject_verb_agreement_1": { "acc,none": 0.808, "acc_stderr,none": 0.012461592646659976, "alias": " - blimp_irregular_plural_subject_verb_agreement_1" }, "blimp_irregular_plural_subject_verb_agreement_2": { "acc,none": 0.766, "acc_stderr,none": 0.01339490288966001, "alias": " - blimp_irregular_plural_subject_verb_agreement_2" }, "blimp_left_branch_island_echo_question": { "acc,none": 0.635, "acc_stderr,none": 0.015231776226264905, "alias": " - blimp_left_branch_island_echo_question" }, "blimp_left_branch_island_simple_question": { "acc,none": 0.543, "acc_stderr,none": 0.015760691590136378, "alias": " - blimp_left_branch_island_simple_question" }, "blimp_matrix_question_npi_licensor_present": { "acc,none": 0.103, "acc_stderr,none": 0.009616833339695796, "alias": " - blimp_matrix_question_npi_licensor_present" }, "blimp_npi_present_1": { "acc,none": 0.545, "acc_stderr,none": 0.01575510149834709, "alias": " - blimp_npi_present_1" }, "blimp_npi_present_2": { "acc,none": 0.572, "acc_stderr,none": 0.01565442624502928, "alias": " - blimp_npi_present_2" }, "blimp_only_npi_licensor_present": { "acc,none": 0.979, "acc_stderr,none": 0.004536472151306522, "alias": " - blimp_only_npi_licensor_present" }, "blimp_only_npi_scope": { "acc,none": 0.482, "acc_stderr,none": 0.015809045699406728, "alias": " - blimp_only_npi_scope" }, "blimp_passive_1": { "acc,none": 0.848, "acc_stderr,none": 0.01135891830347528, "alias": " - blimp_passive_1" }, "blimp_passive_2": { "acc,none": 0.87, "acc_stderr,none": 0.010640169792499364, "alias": " - blimp_passive_2" }, "blimp_principle_A_c_command": { "acc,none": 0.563, "acc_stderr,none": 0.01569322392873038, "alias": " - blimp_principle_A_c_command" }, "blimp_principle_A_case_1": { "acc,none": 1.0, "acc_stderr,none": 0.0, "alias": " - blimp_principle_A_case_1" }, "blimp_principle_A_case_2": { "acc,none": 0.908, "acc_stderr,none": 0.009144376393151103, "alias": " - blimp_principle_A_case_2" }, "blimp_principle_A_domain_1": { "acc,none": 0.986, "acc_stderr,none": 0.0037172325482565964, "alias": " - blimp_principle_A_domain_1" }, "blimp_principle_A_domain_2": { "acc,none": 0.758, "acc_stderr,none": 0.01355063170555596, "alias": " - blimp_principle_A_domain_2" }, "blimp_principle_A_domain_3": { "acc,none": 0.559, "acc_stderr,none": 0.01570877989424268, "alias": " - blimp_principle_A_domain_3" }, "blimp_principle_A_reconstruction": { "acc,none": 0.218, "acc_stderr,none": 0.013063179040595283, "alias": " - blimp_principle_A_reconstruction" }, "blimp_regular_plural_subject_verb_agreement_1": { "acc,none": 0.871, "acc_stderr,none": 0.010605256784796594, "alias": " - blimp_regular_plural_subject_verb_agreement_1" }, "blimp_regular_plural_subject_verb_agreement_2": { "acc,none": 0.755, "acc_stderr,none": 0.01360735683959812, "alias": " - blimp_regular_plural_subject_verb_agreement_2" }, "blimp_sentential_negation_npi_licensor_present": { "acc,none": 0.982, "acc_stderr,none": 0.004206387249611467, "alias": " - blimp_sentential_negation_npi_licensor_present" }, "blimp_sentential_negation_npi_scope": { "acc,none": 0.448, "acc_stderr,none": 0.015733516566347833, "alias": " - blimp_sentential_negation_npi_scope" }, "blimp_sentential_subject_island": { "acc,none": 0.339, "acc_stderr,none": 0.014976758771620342, "alias": " - blimp_sentential_subject_island" }, "blimp_superlative_quantifiers_1": { "acc,none": 0.616, "acc_stderr,none": 0.015387682761897066, "alias": " - blimp_superlative_quantifiers_1" }, "blimp_superlative_quantifiers_2": { "acc,none": 0.773, "acc_stderr,none": 0.01325317496476391, "alias": " - blimp_superlative_quantifiers_2" }, "blimp_tough_vs_raising_1": { "acc,none": 0.551, "acc_stderr,none": 0.015736792768752023, "alias": " - blimp_tough_vs_raising_1" }, "blimp_tough_vs_raising_2": { "acc,none": 0.785, "acc_stderr,none": 0.01299784381903181, "alias": " - blimp_tough_vs_raising_2" }, "blimp_transitive": { "acc,none": 0.792, "acc_stderr,none": 0.01284137457209693, "alias": " - blimp_transitive" }, "blimp_wh_island": { "acc,none": 0.619, "acc_stderr,none": 0.015364734787007436, "alias": " - blimp_wh_island" }, "blimp_wh_questions_object_gap": { "acc,none": 0.734, "acc_stderr,none": 0.013979965645145151, "alias": " - blimp_wh_questions_object_gap" }, "blimp_wh_questions_subject_gap": { "acc,none": 0.917, "acc_stderr,none": 0.008728527206074794, "alias": " - blimp_wh_questions_subject_gap" }, "blimp_wh_questions_subject_gap_long_distance": { "acc,none": 0.896, "acc_stderr,none": 0.009658016218524298, "alias": " - blimp_wh_questions_subject_gap_long_distance" }, "blimp_wh_vs_that_no_gap": { "acc,none": 0.963, "acc_stderr,none": 0.005972157622389644, "alias": " - blimp_wh_vs_that_no_gap" }, "blimp_wh_vs_that_no_gap_long_distance": { "acc,none": 0.982, "acc_stderr,none": 0.004206387249611444, "alias": " - blimp_wh_vs_that_no_gap_long_distance" }, "blimp_wh_vs_that_with_gap": { "acc,none": 0.418, "acc_stderr,none": 0.015605111967541944, "alias": " - blimp_wh_vs_that_with_gap" }, "blimp_wh_vs_that_with_gap_long_distance": { "acc,none": 0.084, "acc_stderr,none": 0.008776162089491134, "alias": " - blimp_wh_vs_that_with_gap_long_distance" } }, "groups": { "blimp": { "acc,none": 0.7272835820895525, "acc_stderr,none": 0.19867610993066823, "alias": "blimp" } }, "configs": { "blimp_adjunct_island": { "task": "blimp_adjunct_island", "group": "blimp", "dataset_path": "blimp", "dataset_name": "adjunct_island", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_anaphor_gender_agreement": { "task": "blimp_anaphor_gender_agreement", "group": "blimp", "dataset_path": "blimp", "dataset_name": "anaphor_gender_agreement", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_anaphor_number_agreement": { "task": "blimp_anaphor_number_agreement", "group": "blimp", "dataset_path": "blimp", "dataset_name": "anaphor_number_agreement", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_animate_subject_passive": { "task": "blimp_animate_subject_passive", "group": "blimp", "dataset_path": "blimp", "dataset_name": "animate_subject_passive", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_animate_subject_trans": { "task": "blimp_animate_subject_trans", "group": "blimp", "dataset_path": "blimp", "dataset_name": "animate_subject_trans", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_causative": { "task": "blimp_causative", "group": "blimp", "dataset_path": "blimp", "dataset_name": "causative", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_complex_NP_island": { "task": "blimp_complex_NP_island", "group": "blimp", "dataset_path": "blimp", "dataset_name": "complex_NP_island", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_coordinate_structure_constraint_complex_left_branch": { "task": "blimp_coordinate_structure_constraint_complex_left_branch", "group": "blimp", "dataset_path": "blimp", "dataset_name": "coordinate_structure_constraint_complex_left_branch", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_coordinate_structure_constraint_object_extraction": { "task": "blimp_coordinate_structure_constraint_object_extraction", "group": "blimp", "dataset_path": "blimp", "dataset_name": "coordinate_structure_constraint_object_extraction", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_1": { "task": "blimp_determiner_noun_agreement_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_2": { "task": "blimp_determiner_noun_agreement_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_irregular_1": { "task": "blimp_determiner_noun_agreement_irregular_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_irregular_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_irregular_2": { "task": "blimp_determiner_noun_agreement_irregular_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_irregular_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_with_adj_2": { "task": "blimp_determiner_noun_agreement_with_adj_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_with_adj_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_with_adj_irregular_1": { "task": "blimp_determiner_noun_agreement_with_adj_irregular_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_with_adj_irregular_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_with_adj_irregular_2": { "task": "blimp_determiner_noun_agreement_with_adj_irregular_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_with_adj_irregular_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_determiner_noun_agreement_with_adjective_1": { "task": "blimp_determiner_noun_agreement_with_adjective_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "determiner_noun_agreement_with_adjective_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_distractor_agreement_relational_noun": { "task": "blimp_distractor_agreement_relational_noun", "group": "blimp", "dataset_path": "blimp", "dataset_name": "distractor_agreement_relational_noun", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_distractor_agreement_relative_clause": { "task": "blimp_distractor_agreement_relative_clause", "group": "blimp", "dataset_path": "blimp", "dataset_name": "distractor_agreement_relative_clause", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_drop_argument": { "task": "blimp_drop_argument", "group": "blimp", "dataset_path": "blimp", "dataset_name": "drop_argument", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_ellipsis_n_bar_1": { "task": "blimp_ellipsis_n_bar_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "ellipsis_n_bar_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_ellipsis_n_bar_2": { "task": "blimp_ellipsis_n_bar_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "ellipsis_n_bar_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_existential_there_object_raising": { "task": "blimp_existential_there_object_raising", "group": "blimp", "dataset_path": "blimp", "dataset_name": "existential_there_object_raising", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_existential_there_quantifiers_1": { "task": "blimp_existential_there_quantifiers_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "existential_there_quantifiers_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_existential_there_quantifiers_2": { "task": "blimp_existential_there_quantifiers_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "existential_there_quantifiers_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_existential_there_subject_raising": { "task": "blimp_existential_there_subject_raising", "group": "blimp", "dataset_path": "blimp", "dataset_name": "existential_there_subject_raising", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_expletive_it_object_raising": { "task": "blimp_expletive_it_object_raising", "group": "blimp", "dataset_path": "blimp", "dataset_name": "expletive_it_object_raising", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_inchoative": { "task": "blimp_inchoative", "group": "blimp", "dataset_path": "blimp", "dataset_name": "inchoative", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_intransitive": { "task": "blimp_intransitive", "group": "blimp", "dataset_path": "blimp", "dataset_name": "intransitive", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_irregular_past_participle_adjectives": { "task": "blimp_irregular_past_participle_adjectives", "group": "blimp", "dataset_path": "blimp", "dataset_name": "irregular_past_participle_adjectives", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_irregular_past_participle_verbs": { "task": "blimp_irregular_past_participle_verbs", "group": "blimp", "dataset_path": "blimp", "dataset_name": "irregular_past_participle_verbs", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_irregular_plural_subject_verb_agreement_1": { "task": "blimp_irregular_plural_subject_verb_agreement_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "irregular_plural_subject_verb_agreement_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_irregular_plural_subject_verb_agreement_2": { "task": "blimp_irregular_plural_subject_verb_agreement_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "irregular_plural_subject_verb_agreement_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_left_branch_island_echo_question": { "task": "blimp_left_branch_island_echo_question", "group": "blimp", "dataset_path": "blimp", "dataset_name": "left_branch_island_echo_question", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_left_branch_island_simple_question": { "task": "blimp_left_branch_island_simple_question", "group": "blimp", "dataset_path": "blimp", "dataset_name": "left_branch_island_simple_question", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_matrix_question_npi_licensor_present": { "task": "blimp_matrix_question_npi_licensor_present", "group": "blimp", "dataset_path": "blimp", "dataset_name": "matrix_question_npi_licensor_present", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_npi_present_1": { "task": "blimp_npi_present_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "npi_present_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_npi_present_2": { "task": "blimp_npi_present_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "npi_present_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_only_npi_licensor_present": { "task": "blimp_only_npi_licensor_present", "group": "blimp", "dataset_path": "blimp", "dataset_name": "only_npi_licensor_present", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_only_npi_scope": { "task": "blimp_only_npi_scope", "group": "blimp", "dataset_path": "blimp", "dataset_name": "only_npi_scope", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_passive_1": { "task": "blimp_passive_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "passive_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_passive_2": { "task": "blimp_passive_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "passive_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_c_command": { "task": "blimp_principle_A_c_command", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_c_command", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_case_1": { "task": "blimp_principle_A_case_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_case_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_case_2": { "task": "blimp_principle_A_case_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_case_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_domain_1": { "task": "blimp_principle_A_domain_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_domain_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_domain_2": { "task": "blimp_principle_A_domain_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_domain_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_domain_3": { "task": "blimp_principle_A_domain_3", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_domain_3", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_principle_A_reconstruction": { "task": "blimp_principle_A_reconstruction", "group": "blimp", "dataset_path": "blimp", "dataset_name": "principle_A_reconstruction", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_regular_plural_subject_verb_agreement_1": { "task": "blimp_regular_plural_subject_verb_agreement_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "regular_plural_subject_verb_agreement_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_regular_plural_subject_verb_agreement_2": { "task": "blimp_regular_plural_subject_verb_agreement_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "regular_plural_subject_verb_agreement_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_sentential_negation_npi_licensor_present": { "task": "blimp_sentential_negation_npi_licensor_present", "group": "blimp", "dataset_path": "blimp", "dataset_name": "sentential_negation_npi_licensor_present", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_sentential_negation_npi_scope": { "task": "blimp_sentential_negation_npi_scope", "group": "blimp", "dataset_path": "blimp", "dataset_name": "sentential_negation_npi_scope", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_sentential_subject_island": { "task": "blimp_sentential_subject_island", "group": "blimp", "dataset_path": "blimp", "dataset_name": "sentential_subject_island", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_superlative_quantifiers_1": { "task": "blimp_superlative_quantifiers_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "superlative_quantifiers_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_superlative_quantifiers_2": { "task": "blimp_superlative_quantifiers_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "superlative_quantifiers_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_tough_vs_raising_1": { "task": "blimp_tough_vs_raising_1", "group": "blimp", "dataset_path": "blimp", "dataset_name": "tough_vs_raising_1", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_tough_vs_raising_2": { "task": "blimp_tough_vs_raising_2", "group": "blimp", "dataset_path": "blimp", "dataset_name": "tough_vs_raising_2", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_transitive": { "task": "blimp_transitive", "group": "blimp", "dataset_path": "blimp", "dataset_name": "transitive", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_island": { "task": "blimp_wh_island", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_island", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_questions_object_gap": { "task": "blimp_wh_questions_object_gap", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_questions_object_gap", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_questions_subject_gap": { "task": "blimp_wh_questions_subject_gap", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_questions_subject_gap", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_questions_subject_gap_long_distance": { "task": "blimp_wh_questions_subject_gap_long_distance", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_questions_subject_gap_long_distance", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_vs_that_no_gap": { "task": "blimp_wh_vs_that_no_gap", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_vs_that_no_gap", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_vs_that_no_gap_long_distance": { "task": "blimp_wh_vs_that_no_gap_long_distance", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_vs_that_no_gap_long_distance", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_vs_that_with_gap": { "task": "blimp_wh_vs_that_with_gap", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_vs_that_with_gap", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } }, "blimp_wh_vs_that_with_gap_long_distance": { "task": "blimp_wh_vs_that_with_gap_long_distance", "group": "blimp", "dataset_path": "blimp", "dataset_name": "wh_vs_that_with_gap_long_distance", "validation_split": "train", "doc_to_text": "", "doc_to_target": 0, "doc_to_choice": "{{[sentence_good, sentence_bad]}}", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "acc" } ], "output_type": "multiple_choice", "repeats": 1, "should_decontaminate": true, "doc_to_decontamination_query": "{{sentence_good}} {{sentence_bad}}", "metadata": { "version": 1.0 } } }, "versions": { "blimp": "N/A", "blimp_adjunct_island": 1.0, "blimp_anaphor_gender_agreement": 1.0, "blimp_anaphor_number_agreement": 1.0, "blimp_animate_subject_passive": 1.0, "blimp_animate_subject_trans": 1.0, "blimp_causative": 1.0, "blimp_complex_NP_island": 1.0, "blimp_coordinate_structure_constraint_complex_left_branch": 1.0, "blimp_coordinate_structure_constraint_object_extraction": 1.0, "blimp_determiner_noun_agreement_1": 1.0, "blimp_determiner_noun_agreement_2": 1.0, "blimp_determiner_noun_agreement_irregular_1": 1.0, "blimp_determiner_noun_agreement_irregular_2": 1.0, "blimp_determiner_noun_agreement_with_adj_2": 1.0, "blimp_determiner_noun_agreement_with_adj_irregular_1": 1.0, "blimp_determiner_noun_agreement_with_adj_irregular_2": 1.0, "blimp_determiner_noun_agreement_with_adjective_1": 1.0, "blimp_distractor_agreement_relational_noun": 1.0, "blimp_distractor_agreement_relative_clause": 1.0, "blimp_drop_argument": 1.0, "blimp_ellipsis_n_bar_1": 1.0, "blimp_ellipsis_n_bar_2": 1.0, "blimp_existential_there_object_raising": 1.0, "blimp_existential_there_quantifiers_1": 1.0, "blimp_existential_there_quantifiers_2": 1.0, "blimp_existential_there_subject_raising": 1.0, "blimp_expletive_it_object_raising": 1.0, "blimp_inchoative": 1.0, "blimp_intransitive": 1.0, "blimp_irregular_past_participle_adjectives": 1.0, "blimp_irregular_past_participle_verbs": 1.0, "blimp_irregular_plural_subject_verb_agreement_1": 1.0, "blimp_irregular_plural_subject_verb_agreement_2": 1.0, "blimp_left_branch_island_echo_question": 1.0, "blimp_left_branch_island_simple_question": 1.0, "blimp_matrix_question_npi_licensor_present": 1.0, "blimp_npi_present_1": 1.0, "blimp_npi_present_2": 1.0, "blimp_only_npi_licensor_present": 1.0, "blimp_only_npi_scope": 1.0, "blimp_passive_1": 1.0, "blimp_passive_2": 1.0, "blimp_principle_A_c_command": 1.0, "blimp_principle_A_case_1": 1.0, "blimp_principle_A_case_2": 1.0, "blimp_principle_A_domain_1": 1.0, "blimp_principle_A_domain_2": 1.0, "blimp_principle_A_domain_3": 1.0, "blimp_principle_A_reconstruction": 1.0, "blimp_regular_plural_subject_verb_agreement_1": 1.0, "blimp_regular_plural_subject_verb_agreement_2": 1.0, "blimp_sentential_negation_npi_licensor_present": 1.0, "blimp_sentential_negation_npi_scope": 1.0, "blimp_sentential_subject_island": 1.0, "blimp_superlative_quantifiers_1": 1.0, "blimp_superlative_quantifiers_2": 1.0, "blimp_tough_vs_raising_1": 1.0, "blimp_tough_vs_raising_2": 1.0, "blimp_transitive": 1.0, "blimp_wh_island": 1.0, "blimp_wh_questions_object_gap": 1.0, "blimp_wh_questions_subject_gap": 1.0, "blimp_wh_questions_subject_gap_long_distance": 1.0, "blimp_wh_vs_that_no_gap": 1.0, "blimp_wh_vs_that_no_gap_long_distance": 1.0, "blimp_wh_vs_that_with_gap": 1.0, "blimp_wh_vs_that_with_gap_long_distance": 1.0 }, "n-shot": { "blimp": 0, "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", "model_args": "pretrained=/home/bastian/Dokumente/tweenie_llamas/models/final_20", "batch_size": 1, "batch_sizes": [], "device": "cuda", "use_cache": null, "limit": null, "bootstrap_iters": 100000, "gen_kwargs": null }, "git_hash": null }