This is OpenLLaMA 3B V2 finetuned on LIMA(ShareGPT format) for 2 epochs.
Prompt template:
### HUMAN:
{prompt}
### RESPONSE:
<leave a newline for the model to answer>
GGUF quantizations available here.
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 40.18 |
| ARC (25-shot) | 40.36 |
| HellaSwag (10-shot) | 72.0 |
| MMLU (5-shot) | 26.43 |
| TruthfulQA (0-shot) | 36.11 |
| Winogrande (5-shot) | 65.67 |
| GSM8K (5-shot) | 0.53 |
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 40.18 |
| AI2 Reasoning Challenge (25-Shot) | 40.36 |
| HellaSwag (10-Shot) | 72.00 |
| MMLU (5-Shot) | 26.43 |
| TruthfulQA (0-shot) | 36.11 |
| Winogrande (5-shot) | 65.67 |
| GSM8k (5-shot) | 0.53 |
- Downloads last month
- 3,039
Dataset used to train acrastt/Bean-3B
Spaces using acrastt/Bean-3B 14
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard40.360
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard72.000
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard26.430
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard36.110
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard65.670
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard0.530