Update README.md
Browse files
README.md
CHANGED
|
@@ -11,6 +11,17 @@ license: apache-2.0
|
|
| 11 |
language:
|
| 12 |
- en
|
| 13 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
|
| 15 |
# Uploaded model
|
| 16 |
|
|
|
|
| 11 |
language:
|
| 12 |
- en
|
| 13 |
---
|
| 14 |
+
# PyRe is Experimental
|
| 15 |
+
|
| 16 |
+
Please note that this model is a WIP experiment into GPRO fine tuning on Python code problems for reasoning. The performance of this model varies greatly depending on task, prompt and parameters.
|
| 17 |
+
|
| 18 |
+
I recommend a very low temperature, like 0.1. You may also see more consistent results by encouraging the use of <think> and <answer> tags in the system prompt.
|
| 19 |
+
|
| 20 |
+
### Example System Prompt
|
| 21 |
+
```
|
| 22 |
+
Think through complex problems carefully, before giving the user your final answer. Use <think> and </think> to encapsulate your thoughts.
|
| 23 |
+
```
|
| 24 |
+
|
| 25 |
|
| 26 |
# Uploaded model
|
| 27 |
|