Starling-LM-7B-alpha-Nous.md

Model	AGIEval	GPT4All	TruthfulQA	Bigbench	Average
Starling-LM-7B-alpha	42.06	72.72	47.33	42.53	51.16

AGIEval

Task	Version	Metric	Value		Stderr
agieval_aqua_rat	0	acc	24.80	±	2.72
		acc_norm	25.98	±	2.76
agieval_logiqa_en	0	acc	38.25	±	1.91
		acc_norm	39.78	±	1.92
agieval_lsat_ar	0	acc	25.22	±	2.87
		acc_norm	23.48	±	2.80
agieval_lsat_lr	0	acc	48.43	±	2.22
		acc_norm	45.69	±	2.21
agieval_lsat_rc	0	acc	60.22	±	2.99
		acc_norm	57.62	±	3.02
agieval_sat_en	0	acc	76.21	±	2.97
		acc_norm	72.82	±	3.11
agieval_sat_en_without_passage	0	acc	41.26	±	3.44
		acc_norm	38.35	±	3.40
agieval_sat_math	0	acc	39.55	±	3.30
		acc_norm	32.73	±	3.17

Average: 42.06%

GPT4All

Task	Version	Metric	Value		Stderr
arc_challenge	0	acc	57.08	±	1.45
		acc_norm	60.24	±	1.43
arc_easy	0	acc	83.59	±	0.76
		acc_norm	81.90	±	0.79
boolq	1	acc	86.79	±	0.59
hellaswag	0	acc	64.26	±	0.48
		acc_norm	82.22	±	0.38
openbookqa	0	acc	29.40	±	2.04
		acc_norm	41.60	±	2.21
piqa	0	acc	81.56	±	0.90
		acc_norm	83.57	±	0.86
winogrande	0	acc	72.69	±	1.25

Average: 72.72%

TruthfulQA

Task	Version	Metric	Value		Stderr
truthfulqa_mc	1	mc1	31.58	±	1.63
		mc2	47.33	±	1.52

Average: 47.33%

Bigbench

Task	Version	Metric	Value		Stderr
bigbench_causal_judgement	0	multiple_choice_grade	62.11	±	3.53
bigbench_date_understanding	0	multiple_choice_grade	64.23	±	2.50
bigbench_disambiguation_qa	0	multiple_choice_grade	50.39	±	3.12
bigbench_geometric_shapes	0	multiple_choice_grade	20.06	±	2.12
		exact_str_match	4.46	±	1.09
bigbench_logical_deduction_five_objects	0	multiple_choice_grade	27.60	±	2.00
bigbench_logical_deduction_seven_objects	0	multiple_choice_grade	18.43	±	1.47
bigbench_logical_deduction_three_objects	0	multiple_choice_grade	50.00	±	2.89
bigbench_movie_recommendation	0	multiple_choice_grade	36.60	±	2.16
bigbench_navigate	0	multiple_choice_grade	50.50	±	1.58
bigbench_reasoning_about_colored_objects	0	multiple_choice_grade	66.95	±	1.05
bigbench_ruin_names	0	multiple_choice_grade	49.11	±	2.36
bigbench_salient_translation_error_detection	0	multiple_choice_grade	15.03	±	1.13
bigbench_snarks	0	multiple_choice_grade	68.51	±	3.46
bigbench_sports_understanding	0	multiple_choice_grade	65.21	±	1.52
bigbench_temporal_sequences	0	multiple_choice_grade	30.90	±	1.46
bigbench_tracking_shuffled_objects_five_objects	0	multiple_choice_grade	23.20	±	1.19
bigbench_tracking_shuffled_objects_seven_objects	0	multiple_choice_grade	16.80	±	0.89
bigbench_tracking_shuffled_objects_three_objects	0	multiple_choice_grade	50.00	±	2.89

Average: 42.53%

Average score: 51.16%