updates `README.md`
Browse files
README.md
CHANGED
|
@@ -44,9 +44,14 @@ you can check our [github](https://github.com/zai-org/GLM-4.5) for more detail.
|
|
| 44 |
|
| 45 |
## Recommended Evaluation Parameters
|
| 46 |
|
| 47 |
-
For general evaluations, we recommend using a **sampling temperature of 1.0**.
|
| 48 |
|
| 49 |
For **code-related evaluation tasks** (such as LCB), it is further recommended to set:
|
| 50 |
|
| 51 |
- `top_p = 0.95`
|
| 52 |
-
- `top_k = 40`
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
## Recommended Evaluation Parameters
|
| 46 |
|
| 47 |
+
For general evaluations, we recommend using a **sampling temperature of 1.0**.
|
| 48 |
|
| 49 |
For **code-related evaluation tasks** (such as LCB), it is further recommended to set:
|
| 50 |
|
| 51 |
- `top_p = 0.95`
|
| 52 |
+
- `top_k = 40`
|
| 53 |
+
|
| 54 |
+
## Evaluation
|
| 55 |
+
|
| 56 |
+
- For tool-integrated reasoning, please refer to [this doc](https://github.com/zai-org/GLM-4.5/blob/main/resources/glm_4.6_tir_guide.md).
|
| 57 |
+
- For search benchmark, we design a specific format for searching toolcall in thinking mode to support search agent, please refer to [this](https://github.com/zai-org/GLM-4.5/blob/main/resources/trajectory_search.json). for the detailed template.
|