Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# OpenRubrics/RubricRM-8B-Judge
|
| 2 |
+
|
| 3 |
+
Finetuned checkpoint for rubric-based reward modeling / judging.
|
| 4 |
+
|
| 5 |
+
## Usage
|
| 6 |
+
```python
|
| 7 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 8 |
+
model_id = "OpenRubrics/RubricRM-8B-Judge"
|
| 9 |
+
tok = AutoTokenizer.from_pretrained(model_id, use_fast=True)
|
| 10 |
+
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto")
|
| 11 |
+
```
|
| 12 |
+
|
| 13 |
+
## Notes
|
| 14 |
+
- Format: Transformers-compatible (config/tokenizer/weights).
|
| 15 |
+
- Base: Qwen3 8B (Rubric-Aware Judge).
|
| 16 |
+
- Files tracked with Git LFS.
|