RQwen v0.1
Short info
- Developed by: ehristoforu
- Base model: Qwen/Qwen2.5-14B-Instruct
- Type model: Qwen2 Instruct (ChatML)
- Languages: English, Russian
- Features: reflection tuning, logic and deep work with context
- Trained with: Unsloth (Transformers SFT)
- License: Apache-2.0
GGUF format: coming soon...
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 32.48 |
| IFEval (0-Shot) | 76.25 |
| BBH (3-Shot) | 48.49 |
| MATH Lvl 5 (4-Shot) | 2.95 |
| GPQA (0-shot) | 10.07 |
| MuSR (0-shot) | 10.44 |
| MMLU-PRO (5-shot) | 46.69 |
- Downloads last month
- 77
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for ehristoforu/RQwen-v0.1
Base model
Qwen/Qwen2.5-14B
Finetuned
Qwen/Qwen2.5-14B-Instruct
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard76.250
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard48.490
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard2.950
- acc_norm on GPQA (0-shot)Open LLM Leaderboard10.070
- acc_norm on MuSR (0-shot)Open LLM Leaderboard10.440
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard46.690