Initial model card submission for Open LLM Leaderboard v2
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- reasoning
|
| 7 |
+
- grpo
|
| 8 |
+
- reinforcement-learning
|
| 9 |
+
- leaderboard
|
| 10 |
+
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
|
| 11 |
+
pipeline_tag: text-generation
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# SAM-AI-Reasoning-14B
|
| 15 |
+
|
| 16 |
+
Official submission of **SAM-AI Reasoning Engine** trained with Group Relative Policy Optimization (GRPO).
|
| 17 |
+
|
| 18 |
+
## Evaluation
|
| 19 |
+
Submitted for independent verification on the **Hugging Face Open LLM Leaderboard v2**:
|
| 20 |
+
- IFEval (Instruction Following)
|
| 21 |
+
- BBH (Big Bench Hard)
|
| 22 |
+
- MATH-Hard (Advanced Competition Mathematics)
|
| 23 |
+
- GPQA Diamond (PhD-level Science)
|
| 24 |
+
- MuSR (Multi-step Soft Reasoning)
|
| 25 |
+
- MMLU-Pro (Professional Knowledge)
|