Skip to content

Instantly share code, notes, and snippets.

@CultriX-Github
Created April 14, 2024 10:34
Show Gist options
  • Save CultriX-Github/02bd1b4a7af690caa071f82b3ee73bff to your computer and use it in GitHub Desktop.
Save CultriX-Github/02bd1b4a7af690caa071f82b3ee73bff to your computer and use it in GitHub Desktop.
Model AGIEval GPT4All TruthfulQA Bigbench Average
MonaCeption-7B-SLERP-DPO 45.57 76.99 78.98 50.19 62.93

AGIEval

Task Version Metric Value Stderr
agieval_aqua_rat 0 acc 27.17 ± 2.80
acc_norm 25.20 ± 2.73
agieval_logiqa_en 0 acc 38.86 ± 1.91
acc_norm 38.71 ± 1.91
agieval_lsat_ar 0 acc 24.78 ± 2.85
acc_norm 24.35 ± 2.84
agieval_lsat_lr 0 acc 53.14 ± 2.21
acc_norm 54.51 ± 2.21
agieval_lsat_rc 0 acc 68.03 ± 2.85
acc_norm 67.66 ± 2.86
agieval_sat_en 0 acc 77.18 ± 2.93
acc_norm 78.16 ± 2.89
agieval_sat_en_without_passage 0 acc 43.20 ± 3.46
acc_norm 44.17 ± 3.47
agieval_sat_math 0 acc 33.64 ± 3.19
acc_norm 31.82 ± 3.15

Average: 45.57%

GPT4All

Task Version Metric Value Stderr
arc_challenge 0 acc 66.55 ± 1.38
acc_norm 67.41 ± 1.37
arc_easy 0 acc 86.24 ± 0.71
acc_norm 80.05 ± 0.82
boolq 1 acc 86.88 ± 0.59
hellaswag 0 acc 69.17 ± 0.46
acc_norm 87.34 ± 0.33
openbookqa 0 acc 39.00 ± 2.18
acc_norm 49.80 ± 2.24
piqa 0 acc 83.24 ± 0.87
acc_norm 85.53 ± 0.82
winogrande 0 acc 81.93 ± 1.08

Average: 76.99%

TruthfulQA

Task Version Metric Value Stderr
truthfulqa_mc 1 mc1 62.55 ± 1.69
mc2 78.98 ± 1.36

Average: 78.98%

Bigbench

Task Version Metric Value Stderr
bigbench_causal_judgement 0 multiple_choice_grade 59.47 ± 3.57
bigbench_date_understanding 0 multiple_choice_grade 62.60 ± 2.52
bigbench_disambiguation_qa 0 multiple_choice_grade 56.20 ± 3.09
bigbench_geometric_shapes 0 multiple_choice_grade 23.68 ± 2.25
exact_str_match 0.84 ± 0.48
bigbench_logical_deduction_five_objects 0 multiple_choice_grade 33.80 ± 2.12
bigbench_logical_deduction_seven_objects 0 multiple_choice_grade 22.57 ± 1.58
bigbench_logical_deduction_three_objects 0 multiple_choice_grade 61.00 ± 2.82
bigbench_movie_recommendation 0 multiple_choice_grade 54.80 ± 2.23
bigbench_navigate 0 multiple_choice_grade 56.80 ± 1.57
bigbench_reasoning_about_colored_objects 0 multiple_choice_grade 69.30 ± 1.03
bigbench_ruin_names 0 multiple_choice_grade 55.13 ± 2.35
bigbench_salient_translation_error_detection 0 multiple_choice_grade 41.18 ± 1.56
bigbench_snarks 0 multiple_choice_grade 71.82 ± 3.35
bigbench_sports_understanding 0 multiple_choice_grade 75.25 ± 1.37
bigbench_temporal_sequences 0 multiple_choice_grade 55.70 ± 1.57
bigbench_tracking_shuffled_objects_five_objects 0 multiple_choice_grade 23.52 ± 1.20
bigbench_tracking_shuffled_objects_seven_objects 0 multiple_choice_grade 19.60 ± 0.95
bigbench_tracking_shuffled_objects_three_objects 0 multiple_choice_grade 61.00 ± 2.82

Average: 50.19%

Average score: 62.93%

Elapsed time: 03:30:28

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