Performance LLMs - Base Models
Collection
22 items
•
Updated
•
6
This is a llamafied qwen-14b for compatibility with the llama software ecosystem.
I used this script to make the model and used the tokenizer of CausalLM, as suggested in the comments of the script.
https://github.com/hiyouga/LLaMA-Factory/blob/main/tests/llamafy_qwen.py
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 63.09 |
| AI2 Reasoning Challenge (25-Shot) | 55.20 |
| HellaSwag (10-Shot) | 82.31 |
| MMLU (5-Shot) | 66.11 |
| TruthfulQA (0-shot) | 45.60 |
| Winogrande (5-shot) | 76.56 |
| GSM8k (5-shot) | 52.77 |
- Downloads last month
- 5,612
Collection including KnutJaegersberg/Qwen-14B-Llamafied
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard55.200
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard82.310
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard66.110
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard45.600
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard76.560
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard52.770