Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,13 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
This is based on ultrachat dataset https://huggingface.co/datasets/stingning/ultrachat
|
| 6 |
+
|
| 7 |
+
I filtered it using the classic "unfiltered" keywords list to remove instances of refusals and bias
|
| 8 |
+
|
| 9 |
+
About 90% of the dataset was removed.
|
| 10 |
+
|
| 11 |
+
What remains (400k conversations) is unlikely to inclinate the model to refuse.
|
| 12 |
+
|
| 13 |
+
I am investigating a less heavy handed approach using dolphin-2.1 to reword any detected refusals.
|