Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,21 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: llama3
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama3
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- zh
|
| 6 |
+
tags:
|
| 7 |
+
- not-for-all-audiences
|
| 8 |
+
datasets:
|
| 9 |
+
- unalignment/toxic-dpo-v0.2
|
| 10 |
+
- shareAI/DPO-zh-en-emoji
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Llama3-70B-Orion-Chinese-SE
|
| 14 |
+
|
| 15 |
+
## 概述
|
| 16 |
+
|
| 17 |
+
本模型是从[failspy/Llama-3-70B-Instruct-abliterated-v3](https://huggingface.co/failspy/Llama-3-70B-Instruct-abliterated-v3)经过DPO微调而来
|
| 18 |
+
|
| 19 |
+
相较于[Orion-zhen/Llama3-70B-Orion-Chinese](https://huggingface.co/Orion-zhen/Llama3-70B-Orion-Chinese), 本模型精简了DPO数据集, 力求对原模型影响最小的情况下获得较好的中文性能和emoji体验
|
| 20 |
+
|
| 21 |
+
本模型的emoji触发概率高于[Orion-zhen/Llama3-70B-Orion-Chinese](https://huggingface.co/Orion-zhen/Llama3-70B-Orion-Chinese), 在其他方面稍逊于后者
|