Add library name, pipeline tag, link to Github (#1)
Browse files- Add library name, pipeline tag, link to Github (cdc752a2cd38b2e92a1ff5eeb97d108780d50713)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,9 +1,14 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
-
|
|
|
|
| 5 |
We change to rope_theta from 10000 to 40000 and extend the context window to 16k.
|
| 6 |
-
Also, we modify the chat_template for the system prompt and add <think>.
|
|
|
|
|
|
|
| 7 |
|
| 8 |
# Citation
|
| 9 |
If you find our model, data, or evaluation code useful, please kindly cite our paper:
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
---
|
| 6 |
+
|
| 7 |
+
The base Qwen2.5-Math-7B model used by LUFFY, described in [Learning to Reason under Off-Policy Guidance](https://huggingface.co/papers/2504.14945).
|
| 8 |
We change to rope_theta from 10000 to 40000 and extend the context window to 16k.
|
| 9 |
+
Also, we modify the chat_template for the system prompt and add <think>.
|
| 10 |
+
|
| 11 |
+
Github: https://github.com/ElliottYan/LUFFY
|
| 12 |
|
| 13 |
# Citation
|
| 14 |
If you find our model, data, or evaluation code useful, please kindly cite our paper:
|