Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model:
|
| 4 |
+
- wzhouad/zephyr-7B-WPO-FP
|
| 5 |
+
- HuggingFaceH4/mistral-7b-sft-beta
|
| 6 |
+
tags:
|
| 7 |
+
- wpo
|
| 8 |
+
- mistral
|
| 9 |
+
- alignment
|
| 10 |
+
datasets:
|
| 11 |
+
- HuggingFaceH4/ultrafeedback_binarized
|
| 12 |
+
pipeline_tag: text-generation
|
| 13 |
+
library_name: transformers
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
following [wzhouad/zephyr-7B-WPO-FP](https://huggingface.co/wzhouad/zephyr-7B-WPO-FP)
|
| 17 |
+
|
| 18 |
+
Transfer original weights from `float32` to `bfloat16` type
|