Skip to content

Instantly share code, notes, and snippets.

@CultriX-Github
Created February 3, 2024 11:18
Show Gist options
  • Save CultriX-Github/3a496688c510c77cfd6a402523c328a0 to your computer and use it in GitHub Desktop.
Save CultriX-Github/3a496688c510c77cfd6a402523c328a0 to your computer and use it in GitHub Desktop.
Model AGIEval GPT4All TruthfulQA Bigbench Average
Subnet6Model3 38.86 70.83 55.86 38.09 50.91

AGIEval

Task Version Metric Value Stderr
agieval_aqua_rat 0 acc 23.62 ± 2.67
acc_norm 25.59 ± 2.74
agieval_logiqa_en 0 acc 35.33 ± 1.87
acc_norm 36.56 ± 1.89
agieval_lsat_ar 0 acc 22.17 ± 2.75
acc_norm 22.61 ± 2.76
agieval_lsat_lr 0 acc 41.57 ± 2.18
acc_norm 41.96 ± 2.19
agieval_lsat_rc 0 acc 49.07 ± 3.05
acc_norm 46.84 ± 3.05
agieval_sat_en 0 acc 70.39 ± 3.19
acc_norm 66.02 ± 3.31
agieval_sat_en_without_passage 0 acc 41.26 ± 3.44
acc_norm 41.26 ± 3.44
agieval_sat_math 0 acc 38.18 ± 3.28
acc_norm 30.00 ± 3.10

Average: 38.86%

GPT4All

Task Version Metric Value Stderr
arc_challenge 0 acc 51.28 ± 1.46
acc_norm 55.97 ± 1.45
arc_easy 0 acc 81.73 ± 0.79
acc_norm 79.55 ± 0.83
boolq 1 acc 84.25 ± 0.64
hellaswag 0 acc 61.14 ± 0.49
acc_norm 79.71 ± 0.40
openbookqa 0 acc 32.00 ± 2.09
acc_norm 42.80 ± 2.21
piqa 0 acc 80.96 ± 0.92
acc_norm 81.18 ± 0.91
winogrande 0 acc 72.38 ± 1.26

Average: 70.83%

TruthfulQA

Task Version Metric Value Stderr
truthfulqa_mc 1 mc1 39.05 ± 1.71
mc2 55.86 ± 1.51

Average: 55.86%

Bigbench

Task Version Metric Value Stderr
bigbench_causal_judgement 0 multiple_choice_grade 55.26 ± 3.62
bigbench_date_understanding 0 multiple_choice_grade 60.16 ± 2.55
bigbench_disambiguation_qa 0 multiple_choice_grade 34.50 ± 2.97
bigbench_geometric_shapes 0 multiple_choice_grade 20.06 ± 2.12
exact_str_match 15.88 ± 1.93
bigbench_logical_deduction_five_objects 0 multiple_choice_grade 26.80 ± 1.98
bigbench_logical_deduction_seven_objects 0 multiple_choice_grade 19.57 ± 1.50
bigbench_logical_deduction_three_objects 0 multiple_choice_grade 48.67 ± 2.89
bigbench_movie_recommendation 0 multiple_choice_grade 33.60 ± 2.11
bigbench_navigate 0 multiple_choice_grade 50.00 ± 1.58
bigbench_reasoning_about_colored_objects 0 multiple_choice_grade 58.90 ± 1.10
bigbench_ruin_names 0 multiple_choice_grade 36.16 ± 2.27
bigbench_salient_translation_error_detection 0 multiple_choice_grade 12.42 ± 1.04
bigbench_snarks 0 multiple_choice_grade 56.91 ± 3.69
bigbench_sports_understanding 0 multiple_choice_grade 58.11 ± 1.57
bigbench_temporal_sequences 0 multiple_choice_grade 27.30 ± 1.41
bigbench_tracking_shuffled_objects_five_objects 0 multiple_choice_grade 22.16 ± 1.18
bigbench_tracking_shuffled_objects_seven_objects 0 multiple_choice_grade 16.40 ± 0.89
bigbench_tracking_shuffled_objects_three_objects 0 multiple_choice_grade 48.67 ± 2.89

Average: 38.09%

Average score: 50.91%

Elapsed time: 02:46:32

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment