MyAwesomeModel-best

Best checkpoint selected from the workspace: checkpoints/step_1000 (highest evaluation score).

Evaluation Results (step_1000, 3 decimal places)

#	Benchmark	Score
1	Math Reasoning	0.550
2	Logical Reasoning	0.819
3	Common Sense	0.736
4	Reading Comprehension	0.700
5	Question Answering	0.607
6	Text Classification	0.828
7	Sentiment Analysis	0.792
8	Code Generation	0.650
9	Creative Writing	0.610
10	Dialogue Generation	0.644
11	Summarization	0.767
12	Translation	0.804
13	Knowledge Retrieval	0.676
14	Instruction Following	0.758
15	Safety Evaluation	0.739

Overall weighted score: 0.710