Skip to content

Instantly share code, notes, and snippets.

@mlabonne
Created January 13, 2024 13:01
Show Gist options
  • Save mlabonne/5266df71633f982dd7fe3a085e9c8829 to your computer and use it in GitHub Desktop.
Save mlabonne/5266df71633f982dd7fe3a085e9c8829 to your computer and use it in GitHub Desktop.
Model AGIEval GPT4All TruthfulQA Bigbench Average
phixtral-4x2_8-gates-poc 31.78 70.22 47.01 37.02 46.51

AGIEval

Task Version Metric Value Stderr
agieval_aqua_rat 0 acc 22.44 ± 2.62
acc_norm 22.83 ± 2.64
agieval_logiqa_en 0 acc 28.88 ± 1.78
acc_norm 29.34 ± 1.79
agieval_lsat_ar 0 acc 22.17 ± 2.75
acc_norm 21.74 ± 2.73
agieval_lsat_lr 0 acc 31.57 ± 2.06
acc_norm 27.84 ± 1.99
agieval_lsat_rc 0 acc 38.29 ± 2.97
acc_norm 33.09 ± 2.87
agieval_sat_en 0 acc 60.19 ± 3.42
acc_norm 54.37 ± 3.48
agieval_sat_en_without_passage 0 acc 40.78 ± 3.43
acc_norm 35.92 ± 3.35
agieval_sat_math 0 acc 30.00 ± 3.10
acc_norm 29.09 ± 3.07

Average: 31.78%

GPT4All

Task Version Metric Value Stderr
arc_challenge 0 acc 50.51 ± 1.46
acc_norm 52.47 ± 1.46
arc_easy 0 acc 79.84 ± 0.82
acc_norm 76.77 ± 0.87
boolq 1 acc 83.82 ± 0.64
hellaswag 0 acc 55.88 ± 0.50
acc_norm 73.77 ± 0.44
openbookqa 0 acc 36.60 ± 2.16
acc_norm 48.80 ± 2.24
piqa 0 acc 79.22 ± 0.95
acc_norm 79.65 ± 0.94
winogrande 0 acc 76.24 ± 1.20

Average: 70.22%

TruthfulQA

Task Version Metric Value Stderr
truthfulqa_mc 1 mc1 33.78 ± 1.66
mc2 47.01 ± 1.52

Average: 47.01%

Bigbench

Task Version Metric Value Stderr
bigbench_causal_judgement 0 multiple_choice_grade 56.32 ± 3.61
bigbench_date_understanding 0 multiple_choice_grade 58.54 ± 2.57
bigbench_disambiguation_qa 0 multiple_choice_grade 39.92 ± 3.05
bigbench_geometric_shapes 0 multiple_choice_grade 20.06 ± 2.12
exact_str_match 0.00 ± 0.00
bigbench_logical_deduction_five_objects 0 multiple_choice_grade 24.40 ± 1.92
bigbench_logical_deduction_seven_objects 0 multiple_choice_grade 15.43 ± 1.37
bigbench_logical_deduction_three_objects 0 multiple_choice_grade 41.33 ± 2.85
bigbench_movie_recommendation 0 multiple_choice_grade 44.20 ± 2.22
bigbench_navigate 0 multiple_choice_grade 53.70 ± 1.58
bigbench_reasoning_about_colored_objects 0 multiple_choice_grade 52.35 ± 1.12
bigbench_ruin_names 0 multiple_choice_grade 32.81 ± 2.22
bigbench_salient_translation_error_detection 0 multiple_choice_grade 24.95 ± 1.37
bigbench_snarks 0 multiple_choice_grade 59.67 ± 3.66
bigbench_sports_understanding 0 multiple_choice_grade 53.65 ± 1.59
bigbench_temporal_sequences 0 multiple_choice_grade 16.40 ± 1.17
bigbench_tracking_shuffled_objects_five_objects 0 multiple_choice_grade 18.88 ± 1.11
bigbench_tracking_shuffled_objects_seven_objects 0 multiple_choice_grade 12.34 ± 0.79
bigbench_tracking_shuffled_objects_three_objects 0 multiple_choice_grade 41.33 ± 2.85

Average: 37.02%

Average score: 46.51%

Elapsed time: 02:23:39

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