Update README.md
Browse files
README.md
CHANGED
|
@@ -6,8 +6,6 @@
|
|
| 6 |
# dataset Intruction
|
| 7 |
---
|
| 8 |
**datasets:** \
|
| 9 |
-
- mxz/alpaca_en_zh_ruozhiba_gpt4data \
|
| 10 |
-
- PKU-Alignment/PKU-SafeRLHF \
|
| 11 |
- mxz/CValues_DPO \
|
| 12 |
**language:** \
|
| 13 |
- zh \
|
|
@@ -17,7 +15,7 @@
|
|
| 17 |
**pipeline_tag:** \
|
| 18 |
- text-generation \
|
| 19 |
**tags:** \
|
| 20 |
-
-
|
| 21 |
- fintune \
|
| 22 |
- alignment \
|
| 23 |
- LoRA \
|
|
@@ -42,7 +40,7 @@ Result:
|
|
| 42 |
| ------------------- | ----- | ------ | ------ |
|
| 43 |
| Llama-3-8B | 55.5 | 47.0 | 48.0 |
|
| 44 |
| Llama-3-8B-Instruct | 60.1 | 49.7 | 49.3 |
|
| 45 |
-
| Llama-3-8B-
|
| 46 |
|
| 47 |
- Llama-3-8B evaluation result from [ymcui/Chinese-LLaMA-Alpaca-3](https://github.com/ymcui/Chinese-LLaMA-Alpaca-3)
|
| 48 |
|
|
|
|
| 6 |
# dataset Intruction
|
| 7 |
---
|
| 8 |
**datasets:** \
|
|
|
|
|
|
|
| 9 |
- mxz/CValues_DPO \
|
| 10 |
**language:** \
|
| 11 |
- zh \
|
|
|
|
| 15 |
**pipeline_tag:** \
|
| 16 |
- text-generation \
|
| 17 |
**tags:** \
|
| 18 |
+
- DPO \
|
| 19 |
- fintune \
|
| 20 |
- alignment \
|
| 21 |
- LoRA \
|
|
|
|
| 40 |
| ------------------- | ----- | ------ | ------ |
|
| 41 |
| Llama-3-8B | 55.5 | 47.0 | 48.0 |
|
| 42 |
| Llama-3-8B-Instruct | 60.1 | 49.7 | 49.3 |
|
| 43 |
+
| Llama-3-8B-dpo | 62.2 | 49.9 | 49.4 |
|
| 44 |
|
| 45 |
- Llama-3-8B evaluation result from [ymcui/Chinese-LLaMA-Alpaca-3](https://github.com/ymcui/Chinese-LLaMA-Alpaca-3)
|
| 46 |
|