procedure2012's picture
Finalize model card with evaluation results
ec6a766 verified
|
Raw
History Blame Contribute Delete
200 Bytes
# InstructLite
## Evaluation Results
| Benchmark | Score |
|---|---|
| MMLU | 71.30 |
| GSM8K | 58.03 |
| HumanEval | 44.67 |
| ARC-Challenge | 66.90 |
| HellaSwag | 82.40 |
| **Average** | 64.66 |