lewtun's picture
lewtun HF staff
Upload eval_results/HuggingFaceH4/zephyr-7b-alpha/main/bbh/results_2024-03-28T15-04-26.956255.json with huggingface_hub
561f184 verified
raw
history blame
33.5 kB
{
"config_general": {
"lighteval_sha": "?",
"num_fewshot_seeds": 1,
"override_batch_size": 1,
"max_samples": null,
"job_id": "",
"start_time": 19935.04172753,
"end_time": 20184.797414068,
"total_evaluation_time_secondes": "249.75568653800292",
"model_name": "HuggingFaceH4/zephyr-7b-alpha",
"model_sha": "2ce2d025864af849b3e5029e2ec9d568eeda892d",
"model_dtype": "torch.bfloat16",
"model_size": "13.99 GB",
"config": null
},
"results": {
"harness|bbh:causal_judgment|3": {
"em": 0.48663101604278075,
"em_stderr": 0.03664867131244299,
"qem": 0.48663101604278075,
"qem_stderr": 0.03664867131244299,
"pem": 0.5828877005347594,
"pem_stderr": 0.03615450931140827,
"pqem": 0.5828877005347594,
"pqem_stderr": 0.03615450931140827,
"perfect_em": 0.48663101604278075,
"perfect_em_stderr": 0.03664867131244299
},
"harness|bbh:date_understanding|3": {
"em": 0.4,
"em_stderr": 0.031046021028253254,
"qem": 0.4,
"qem_stderr": 0.031046021028253254,
"pem": 0.4,
"pem_stderr": 0.031046021028253254,
"pqem": 0.528,
"pqem_stderr": 0.03163648953154439,
"perfect_em": 0.4,
"perfect_em_stderr": 0.031046021028253254
},
"harness|bbh:disambiguation_qa|3": {
"em": 0.66,
"em_stderr": 0.03002007360545791,
"qem": 0.66,
"qem_stderr": 0.03002007360545791,
"pem": 0.664,
"pem_stderr": 0.02993325909419152,
"pqem": 0.764,
"pqem_stderr": 0.02690933759495386,
"perfect_em": 0.66,
"perfect_em_stderr": 0.03002007360545791
},
"harness|bbh:geometric_shapes|3": {
"em": 0.236,
"em_stderr": 0.02690933759495382,
"qem": 0.236,
"qem_stderr": 0.02690933759495382,
"pem": 0.236,
"pem_stderr": 0.02690933759495382,
"pqem": 0.236,
"pqem_stderr": 0.02690933759495382,
"perfect_em": 0.236,
"perfect_em_stderr": 0.02690933759495382
},
"harness|bbh:logical_deduction_five_objects|3": {
"em": 0.388,
"em_stderr": 0.030881038748993915,
"qem": 0.388,
"qem_stderr": 0.030881038748993915,
"pem": 0.388,
"pem_stderr": 0.030881038748993915,
"pqem": 0.536,
"pqem_stderr": 0.03160397514522374,
"perfect_em": 0.388,
"perfect_em_stderr": 0.030881038748993915
},
"harness|bbh:logical_deduction_seven_objects|3": {
"em": 0.32,
"em_stderr": 0.029561724955241044,
"qem": 0.32,
"qem_stderr": 0.029561724955241044,
"pem": 0.32,
"pem_stderr": 0.029561724955241044,
"pqem": 0.456,
"pqem_stderr": 0.03156328506121339,
"perfect_em": 0.32,
"perfect_em_stderr": 0.029561724955241044
},
"harness|bbh:logical_deduction_three_objects|3": {
"em": 0.456,
"em_stderr": 0.03156328506121339,
"qem": 0.456,
"qem_stderr": 0.03156328506121339,
"pem": 0.464,
"pem_stderr": 0.03160397514522374,
"pqem": 0.752,
"pqem_stderr": 0.027367497504863548,
"perfect_em": 0.456,
"perfect_em_stderr": 0.03156328506121339
},
"harness|bbh:movie_recommendation|3": {
"em": 0.6144578313253012,
"em_stderr": 0.030906949634000837,
"qem": 0.6144578313253012,
"qem_stderr": 0.030906949634000837,
"pem": 0.6144578313253012,
"pem_stderr": 0.030906949634000837,
"pqem": 0.714859437751004,
"pqem_stderr": 0.028669095281001408,
"perfect_em": 0.6144578313253012,
"perfect_em_stderr": 0.030906949634000837
},
"harness|bbh:navigate|3": {
"em": 0.588,
"em_stderr": 0.031191596026022898,
"qem": 0.588,
"qem_stderr": 0.031191596026022898,
"pem": 0.588,
"pem_stderr": 0.031191596026022898,
"pqem": 0.588,
"pqem_stderr": 0.031191596026022898,
"perfect_em": 0.588,
"perfect_em_stderr": 0.031191596026022898
},
"harness|bbh:reasoning_about_colored_objects|3": {
"em": 0.324,
"em_stderr": 0.02965829492454557,
"qem": 0.324,
"qem_stderr": 0.02965829492454557,
"pem": 0.328,
"pem_stderr": 0.02975239182447538,
"pqem": 0.444,
"pqem_stderr": 0.03148684942554574,
"perfect_em": 0.324,
"perfect_em_stderr": 0.02965829492454557
},
"harness|bbh:ruin_names|3": {
"em": 0.5161290322580645,
"em_stderr": 0.031797681167106,
"qem": 0.5161290322580645,
"qem_stderr": 0.031797681167106,
"pem": 0.5161290322580645,
"pem_stderr": 0.031797681167106,
"pqem": 0.7096774193548387,
"pqem_stderr": 0.028881678193520312,
"perfect_em": 0.5161290322580645,
"perfect_em_stderr": 0.031797681167106
},
"harness|bbh:salient_translation_error_detection|3": {
"em": 0.36,
"em_stderr": 0.030418764025174985,
"qem": 0.36,
"qem_stderr": 0.030418764025174985,
"pem": 0.364,
"pem_stderr": 0.030491555220405562,
"pqem": 0.496,
"pqem_stderr": 0.031685198551199154,
"perfect_em": 0.36,
"perfect_em_stderr": 0.030418764025174985
},
"harness|bbh:snarks|3": {
"em": 0.6797752808988764,
"em_stderr": 0.03506900770722061,
"qem": 0.6797752808988764,
"qem_stderr": 0.03506900770722061,
"pem": 0.6853932584269663,
"pem_stderr": 0.03490336455459108,
"pqem": 0.8089887640449438,
"pqem_stderr": 0.029547046053276592,
"perfect_em": 0.6797752808988764,
"perfect_em_stderr": 0.03506900770722061
},
"harness|bbh:sports_understanding|3": {
"em": 0.74,
"em_stderr": 0.0277973157526443,
"qem": 0.74,
"qem_stderr": 0.0277973157526443,
"pem": 0.772,
"pem_stderr": 0.02658743248726848,
"pqem": 0.772,
"pqem_stderr": 0.02658743248726848,
"perfect_em": 0.74,
"perfect_em_stderr": 0.0277973157526443
},
"harness|bbh:temporal_sequences|3": {
"em": 0.3,
"em_stderr": 0.029040893477575856,
"qem": 0.3,
"qem_stderr": 0.029040893477575856,
"pem": 0.3,
"pem_stderr": 0.029040893477575856,
"pqem": 0.528,
"pqem_stderr": 0.031636489531544396,
"perfect_em": 0.3,
"perfect_em_stderr": 0.029040893477575856
},
"harness|bbh:tracking_shuffled_objects_five_objects|3": {
"em": 0.192,
"em_stderr": 0.02496069198917202,
"qem": 0.192,
"qem_stderr": 0.02496069198917202,
"pem": 0.192,
"pem_stderr": 0.02496069198917202,
"pqem": 0.392,
"pqem_stderr": 0.0309382076204012,
"perfect_em": 0.192,
"perfect_em_stderr": 0.02496069198917202
},
"harness|bbh:tracking_shuffled_objects_seven_objects|3": {
"em": 0.112,
"em_stderr": 0.019985536939171464,
"qem": 0.112,
"qem_stderr": 0.019985536939171464,
"pem": 0.112,
"pem_stderr": 0.019985536939171464,
"pqem": 0.256,
"pqem_stderr": 0.02765710871820491,
"perfect_em": 0.112,
"perfect_em_stderr": 0.019985536939171464
},
"harness|bbh:tracking_shuffled_objects_three_objects|3": {
"em": 0.324,
"em_stderr": 0.029658294924545574,
"qem": 0.324,
"qem_stderr": 0.029658294924545574,
"pem": 0.324,
"pem_stderr": 0.029658294924545574,
"pqem": 0.64,
"pqem_stderr": 0.030418764025174967,
"perfect_em": 0.324,
"perfect_em_stderr": 0.029658294924545574
},
"harness|bbh:_average|3": {
"em": 0.42761073114027903,
"em_stderr": 0.02983973215965202,
"qem": 0.42761073114027903,
"qem_stderr": 0.02983973215965202,
"pem": 0.43615932347472736,
"pem_stderr": 0.02974256967347782,
"pqem": 0.5669118512047526,
"pqem_stderr": 0.030046883203184507,
"perfect_em": 0.42761073114027903,
"perfect_em_stderr": 0.02983973215965202
}
},
"versions": {
"harness|bbh:causal_judgment|3": 0,
"harness|bbh:date_understanding|3": 0,
"harness|bbh:disambiguation_qa|3": 0,
"harness|bbh:geometric_shapes|3": 0,
"harness|bbh:logical_deduction_five_objects|3": 0,
"harness|bbh:logical_deduction_seven_objects|3": 0,
"harness|bbh:logical_deduction_three_objects|3": 0,
"harness|bbh:movie_recommendation|3": 0,
"harness|bbh:navigate|3": 0,
"harness|bbh:reasoning_about_colored_objects|3": 0,
"harness|bbh:ruin_names|3": 0,
"harness|bbh:salient_translation_error_detection|3": 0,
"harness|bbh:snarks|3": 0,
"harness|bbh:sports_understanding|3": 0,
"harness|bbh:temporal_sequences|3": 0,
"harness|bbh:tracking_shuffled_objects_five_objects|3": 0,
"harness|bbh:tracking_shuffled_objects_seven_objects|3": 0,
"harness|bbh:tracking_shuffled_objects_three_objects|3": 0
},
"config_tasks": {
"harness|bbh:causal_judgment": {
"name": "bbh:causal_judgment",
"prompt_function": "bbh_causal_judgment",
"hf_repo": "lukaemon/bbh",
"hf_subset": "causal_judgement",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 187,
"effective_num_docs": 187,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:date_understanding": {
"name": "bbh:date_understanding",
"prompt_function": "bbh_date_understanding",
"hf_repo": "lukaemon/bbh",
"hf_subset": "date_understanding",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:disambiguation_qa": {
"name": "bbh:disambiguation_qa",
"prompt_function": "bbh_disambiguation_qa",
"hf_repo": "lukaemon/bbh",
"hf_subset": "disambiguation_qa",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:geometric_shapes": {
"name": "bbh:geometric_shapes",
"prompt_function": "bbh_geometric_shapes",
"hf_repo": "lukaemon/bbh",
"hf_subset": "geometric_shapes",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:logical_deduction_five_objects": {
"name": "bbh:logical_deduction_five_objects",
"prompt_function": "bbh_logical_deduction_five_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "logical_deduction_five_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:logical_deduction_seven_objects": {
"name": "bbh:logical_deduction_seven_objects",
"prompt_function": "bbh_logical_deduction_seven_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "logical_deduction_seven_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:logical_deduction_three_objects": {
"name": "bbh:logical_deduction_three_objects",
"prompt_function": "bbh_logical_deduction_three_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "logical_deduction_three_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:movie_recommendation": {
"name": "bbh:movie_recommendation",
"prompt_function": "bbh_movie_recommendation",
"hf_repo": "lukaemon/bbh",
"hf_subset": "movie_recommendation",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 249,
"effective_num_docs": 249,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:navigate": {
"name": "bbh:navigate",
"prompt_function": "bbh_navigate",
"hf_repo": "lukaemon/bbh",
"hf_subset": "navigate",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:reasoning_about_colored_objects": {
"name": "bbh:reasoning_about_colored_objects",
"prompt_function": "bbh_reasoning_about_colored_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "reasoning_about_colored_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:ruin_names": {
"name": "bbh:ruin_names",
"prompt_function": "bbh_ruin_names",
"hf_repo": "lukaemon/bbh",
"hf_subset": "ruin_names",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 248,
"effective_num_docs": 248,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:salient_translation_error_detection": {
"name": "bbh:salient_translation_error_detection",
"prompt_function": "bbh_salient_translation_error_detection",
"hf_repo": "lukaemon/bbh",
"hf_subset": "salient_translation_error_detection",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:snarks": {
"name": "bbh:snarks",
"prompt_function": "bbh_snarks",
"hf_repo": "lukaemon/bbh",
"hf_subset": "snarks",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 178,
"effective_num_docs": 178,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:sports_understanding": {
"name": "bbh:sports_understanding",
"prompt_function": "bbh_sports_understanding",
"hf_repo": "lukaemon/bbh",
"hf_subset": "sports_understanding",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:temporal_sequences": {
"name": "bbh:temporal_sequences",
"prompt_function": "bbh_temporal_sequences",
"hf_repo": "lukaemon/bbh",
"hf_subset": "temporal_sequences",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:tracking_shuffled_objects_five_objects": {
"name": "bbh:tracking_shuffled_objects_five_objects",
"prompt_function": "bbh_tracking_shuffled_objects_five_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "tracking_shuffled_objects_five_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:tracking_shuffled_objects_seven_objects": {
"name": "bbh:tracking_shuffled_objects_seven_objects",
"prompt_function": "bbh_tracking_shuffled_objects_seven_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "tracking_shuffled_objects_seven_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
},
"harness|bbh:tracking_shuffled_objects_three_objects": {
"name": "bbh:tracking_shuffled_objects_three_objects",
"prompt_function": "bbh_tracking_shuffled_objects_three_objects",
"hf_repo": "lukaemon/bbh",
"hf_subset": "tracking_shuffled_objects_three_objects",
"metric": [
"exact_match",
"quasi_exact_match",
"prefix_exact_match",
"prefix_quasi_exact_match",
"perfect_exact_match"
],
"hf_avail_splits": [
"test"
],
"evaluation_splits": [
"test"
],
"few_shots_split": null,
"few_shots_select": null,
"generation_size": 20,
"stop_sequence": [
"</s>",
"Q:",
"\n\n"
],
"output_regex": null,
"frozen": false,
"suite": [
"harness"
],
"original_num_docs": 250,
"effective_num_docs": 250,
"trust_dataset": true,
"must_remove_duplicate_docs": null
}
},
"summary_tasks": {
"harness|bbh:causal_judgment|3": {
"hashes": {
"hash_examples": "63218f5ae055ab2b",
"hash_full_prompts": "fa8168f39a475fb0",
"hash_input_tokens": "787f75e06fd43c0d",
"hash_cont_tokens": "f93a64030a534494"
},
"truncated": 187,
"non_truncated": 0,
"padded": 0,
"non_padded": 187,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:date_understanding|3": {
"hashes": {
"hash_examples": "f145c7a06def3c8e",
"hash_full_prompts": "2cceeea606638d49",
"hash_input_tokens": "10c13d6fb8af7c22",
"hash_cont_tokens": "71c24febc632f7e1"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:disambiguation_qa|3": {
"hashes": {
"hash_examples": "19677fd1773f7eb9",
"hash_full_prompts": "d8f1ba70c22ae578",
"hash_input_tokens": "c21a88707f480cab",
"hash_cont_tokens": "cd829d4290e49c3a"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:geometric_shapes|3": {
"hashes": {
"hash_examples": "76c7b11a13cc72a9",
"hash_full_prompts": "52a60ed1d0113b8b",
"hash_input_tokens": "10e113b2cf3fa584",
"hash_cont_tokens": "390c381ef50f61a1"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:logical_deduction_five_objects|3": {
"hashes": {
"hash_examples": "0e958c856332a745",
"hash_full_prompts": "253aa9791c941909",
"hash_input_tokens": "0bc166cab0aed76a",
"hash_cont_tokens": "19038e402d10ac3c"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:logical_deduction_seven_objects|3": {
"hashes": {
"hash_examples": "ab9de25a5eb40d09",
"hash_full_prompts": "aa6117f601cd268e",
"hash_input_tokens": "ab99c78b48e3a0bb",
"hash_cont_tokens": "87ae3cab2dc25553"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:logical_deduction_three_objects|3": {
"hashes": {
"hash_examples": "3c6bf52517714218",
"hash_full_prompts": "1892b050bc7848a4",
"hash_input_tokens": "a720b56aa7c52551",
"hash_cont_tokens": "39f07c97c15c41ad"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:movie_recommendation|3": {
"hashes": {
"hash_examples": "2d9dc4975935d31a",
"hash_full_prompts": "8e00606ed3407167",
"hash_input_tokens": "c825ab1c99245a17",
"hash_cont_tokens": "a9f7db5f085fe85b"
},
"truncated": 249,
"non_truncated": 0,
"padded": 0,
"non_padded": 249,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:navigate|3": {
"hashes": {
"hash_examples": "ba91dcdb9a064255",
"hash_full_prompts": "8d50c5baf1df7aef",
"hash_input_tokens": "f234e6b28ea1fa49",
"hash_cont_tokens": "97a012232529896a"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:reasoning_about_colored_objects|3": {
"hashes": {
"hash_examples": "a6ba328c4c3385d2",
"hash_full_prompts": "3d2441a21c12a960",
"hash_input_tokens": "f3b577892955aa84",
"hash_cont_tokens": "5d5ded07b5969478"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:ruin_names|3": {
"hashes": {
"hash_examples": "2ef28d5f2d4fdd25",
"hash_full_prompts": "ba95caa786f313b1",
"hash_input_tokens": "9954b30d4205604a",
"hash_cont_tokens": "f1dcbf9b74a9b6c3"
},
"truncated": 248,
"non_truncated": 0,
"padded": 0,
"non_padded": 248,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:salient_translation_error_detection|3": {
"hashes": {
"hash_examples": "c13f25ec8ffed496",
"hash_full_prompts": "a8512d174e1cab8f",
"hash_input_tokens": "3e738df24b7eddf8",
"hash_cont_tokens": "53fdd12107a04044"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:snarks|3": {
"hashes": {
"hash_examples": "5f6db7bff7f6f22e",
"hash_full_prompts": "ff91d81466b9041f",
"hash_input_tokens": "21388b09e13d0208",
"hash_cont_tokens": "2ab329328e48f9e5"
},
"truncated": 178,
"non_truncated": 0,
"padded": 0,
"non_padded": 178,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:sports_understanding|3": {
"hashes": {
"hash_examples": "042afbe5d9c1f02d",
"hash_full_prompts": "a59324d9eb37e0f5",
"hash_input_tokens": "0ad41bb8d2290a5b",
"hash_cont_tokens": "3dc39f16a818d665"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:temporal_sequences|3": {
"hashes": {
"hash_examples": "803a05f352eb6afc",
"hash_full_prompts": "1b3971192bf481e7",
"hash_input_tokens": "3051b60940ccceab",
"hash_cont_tokens": "80ba52ec276591f2"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:tracking_shuffled_objects_five_objects|3": {
"hashes": {
"hash_examples": "2bbac6db7ab0d527",
"hash_full_prompts": "7ef4567d2fcf5094",
"hash_input_tokens": "b841310ee5531238",
"hash_cont_tokens": "dd3afecc9035c64f"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:tracking_shuffled_objects_seven_objects|3": {
"hashes": {
"hash_examples": "845caf093ac2b58c",
"hash_full_prompts": "196a0f8712857624",
"hash_input_tokens": "3e738df24b7eddf8",
"hash_cont_tokens": "63e144fd1b5dc44e"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
},
"harness|bbh:tracking_shuffled_objects_three_objects|3": {
"hashes": {
"hash_examples": "9004f14d5a32b9a8",
"hash_full_prompts": "592a03f0518f17b6",
"hash_input_tokens": "19e0ef1dd5ae9d33",
"hash_cont_tokens": "027d23abdf4fc8ab"
},
"truncated": 250,
"non_truncated": 0,
"padded": 0,
"non_padded": 250,
"effective_few_shots": 3.0,
"num_truncated_few_shots": 0
}
},
"summary_general": {
"hashes": {
"hash_examples": "4ff1e3dc5703575d",
"hash_full_prompts": "0d80ce968d89d4ef",
"hash_input_tokens": "72bda1e7aeb34786",
"hash_cont_tokens": "c57fed96739edd70"
},
"truncated": 4362,
"non_truncated": 0,
"padded": 0,
"non_padded": 4362,
"num_truncated_few_shots": 0
}
}