phi-2-OpenHermes-2.5-Nous.md

Model	AGIEval	GPT4All	TruthfulQA	Bigbench	Average
phi-2-OpenHermes-2.5	30.27	71.18	43.87	35.9	45.3

AGIEval

Task	Version	Metric	Value		Stderr
agieval_aqua_rat	0	acc	18.90	±	2.46
		acc_norm	19.69	±	2.50
agieval_logiqa_en	0	acc	28.73	±	1.77
		acc_norm	31.80	±	1.83
agieval_lsat_ar	0	acc	19.13	±	2.60
		acc_norm	19.57	±	2.62
agieval_lsat_lr	0	acc	30.20	±	2.03
		acc_norm	28.04	±	1.99
agieval_lsat_rc	0	acc	37.92	±	2.96
		acc_norm	32.71	±	2.87
agieval_sat_en	0	acc	52.91	±	3.49
		acc_norm	47.57	±	3.49
agieval_sat_en_without_passage	0	acc	39.32	±	3.41
		acc_norm	36.41	±	3.36
agieval_sat_math	0	acc	30.00	±	3.10
		acc_norm	26.36	±	2.98

Average: 30.27%

GPT4All

Task	Version	Metric	Value		Stderr
arc_challenge	0	acc	50.94	±	1.46
		acc_norm	53.24	±	1.46
arc_easy	0	acc	80.77	±	0.81
		acc_norm	78.70	±	0.84
boolq	1	acc	84.22	±	0.64
hellaswag	0	acc	55.94	±	0.50
		acc_norm	73.80	±	0.44
openbookqa	0	acc	38.80	±	2.18
		acc_norm	50.60	±	2.24
piqa	0	acc	79.11	±	0.95
		acc_norm	79.98	±	0.93
winogrande	0	acc	77.74	±	1.17

Average: 71.18%

TruthfulQA

Task	Version	Metric	Value		Stderr
truthfulqa_mc	1	mc1	30.72	±	1.62
		mc2	43.87	±	1.52

Average: 43.87%

Bigbench

Task	Version	Metric	Value		Stderr
bigbench_causal_judgement	0	multiple_choice_grade	61.05	±	3.55
bigbench_date_understanding	0	multiple_choice_grade	59.35	±	2.56
bigbench_disambiguation_qa	0	multiple_choice_grade	40.31	±	3.06
bigbench_geometric_shapes	0	multiple_choice_grade	10.03	±	1.59
		exact_str_match	6.69	±	1.32
bigbench_logical_deduction_five_objects	0	multiple_choice_grade	25.40	±	1.95
bigbench_logical_deduction_seven_objects	0	multiple_choice_grade	16.71	±	1.41
bigbench_logical_deduction_three_objects	0	multiple_choice_grade	41.67	±	2.85
bigbench_movie_recommendation	0	multiple_choice_grade	39.40	±	2.19
bigbench_navigate	0	multiple_choice_grade	50.00	±	1.58
bigbench_reasoning_about_colored_objects	0	multiple_choice_grade	56.10	±	1.11
bigbench_ruin_names	0	multiple_choice_grade	26.12	±	2.08
bigbench_salient_translation_error_detection	0	multiple_choice_grade	24.95	±	1.37
bigbench_snarks	0	multiple_choice_grade	58.56	±	3.67
bigbench_sports_understanding	0	multiple_choice_grade	50.00	±	1.59
bigbench_temporal_sequences	0	multiple_choice_grade	14.60	±	1.12
bigbench_tracking_shuffled_objects_five_objects	0	multiple_choice_grade	17.52	±	1.08
bigbench_tracking_shuffled_objects_seven_objects	0	multiple_choice_grade	12.80	±	0.80
bigbench_tracking_shuffled_objects_three_objects	0	multiple_choice_grade	41.67	±	2.85

Average: 35.9%

Average score: 45.3%

Elapsed time: 01:23:31