Commit
•
ec8f0c1
1
Parent(s):
78ad800
Update README.md (#18)
Browse files- Update README.md (fd1c9d64e09a1142ad20c7dbd6291c814b227f6f)
Co-authored-by: Dan Saattrup Nielsen <[email protected]>
README.md
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
---
|
| 2 |
license: other
|
| 3 |
inference: false
|
|
|
|
| 4 |
---
|
| 5 |
<!-- header start -->
|
| 6 |
<div style="width: 100%;">
|
|
@@ -183,4 +184,4 @@ A preliminary evaluation of the model quality is conducted by creating a set of
|
|
| 183 |
|
| 184 |
## Major updates of weights v1.1
|
| 185 |
- Refactor the tokenization and separator. In Vicuna v1.1, the separator has been changed from `"###"` to the EOS token `"</s>"`. This change makes it easier to determine the generation stop criteria and enables better compatibility with other libraries.
|
| 186 |
-
- Fix the supervised fine-tuning loss computation for better model quality.
|
|
|
|
| 1 |
---
|
| 2 |
license: other
|
| 3 |
inference: false
|
| 4 |
+
pipeline_tag: conversational
|
| 5 |
---
|
| 6 |
<!-- header start -->
|
| 7 |
<div style="width: 100%;">
|
|
|
|
| 184 |
|
| 185 |
## Major updates of weights v1.1
|
| 186 |
- Refactor the tokenization and separator. In Vicuna v1.1, the separator has been changed from `"###"` to the EOS token `"</s>"`. This change makes it easier to determine the generation stop criteria and enables better compatibility with other libraries.
|
| 187 |
+
- Fix the supervised fine-tuning loss computation for better model quality.
|