Update README.md
Browse files
README.md
CHANGED
|
@@ -83,4 +83,11 @@ Issues, questions, or feedback welcome! Particularly interested in:
|
|
| 83 |
- Template compatibility findings
|
| 84 |
- Comparison with other RP models
|
| 85 |
|
| 86 |
-
Feel free to make a post in the Community tab here!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
- Template compatibility findings
|
| 84 |
- Comparison with other RP models
|
| 85 |
|
| 86 |
+
Feel free to make a post in the Community tab here!
|
| 87 |
+
|
| 88 |
+
## Why train on a base model?
|
| 89 |
+
According to this paper: [Base Models Beat Aligned Models at Randomness and Creativity](https://arxiv.org/abs/2505.00047); and to avoid any possible "GPT-isms", I decided to train on a base model.
|
| 90 |
+
Think of it as more mallable clay vs re-shaping something that was already formed to be something else.
|
| 91 |
+
|
| 92 |
+
This is what led to the behavior observed in this model, where the model just legitimately doesn't understand being an "assistant" outside of being a character that is an assistant.
|
| 93 |
+
SO while the model is probably not useful outside of RP, it is also not intended to be.
|