license
| license |
|---|
| llama2 |
ShearedPlats-7b
An experimental finetune of Sheared LLaMA 2.7b with Alpaca-QLoRA (version 2)Datasets
Trained on alpca style datasetsPrompt Template
Uses alpaca style prompt templateOpen LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 36.72 |
| ARC (25-shot) | 42.41 |
| HellaSwag (10-shot) | 72.58 |
| MMLU (5-shot) | 27.52 |
| TruthfulQA (0-shot) | 39.76 |
| Winogrande (5-shot) | 65.9 |
| GSM8K (5-shot) | 1.52 |
| DROP (3-shot) | 7.34 |