Update README.md
Browse files
README.md
CHANGED
@@ -9,8 +9,8 @@ language:
|
|
9 |
|
10 |
This is a Chinese instruction-tuning lora checkpoint based on llama-7B from [this repo's work](https://github.com/Facico/Chinese-Vicuna)
|
11 |
|
12 |
-
We
|
13 |
-
We finetune
|
14 |
|
15 |
You can use it like this:
|
16 |
```python
|
|
|
9 |
|
10 |
This is a Chinese instruction-tuning lora checkpoint based on llama-7B from [this repo's work](https://github.com/Facico/Chinese-Vicuna)
|
11 |
|
12 |
+
We use the 50k Chinese data, which is the combination of [alpaca_chinese_instruction_dataset](https://github.com/hikariming/alpaca_chinese_dataset.git) and the Chinese conversation data from sharegpt-90k data.
|
13 |
+
We finetune the model for 3 epochs use a single 4090 with ctxlen=2048.
|
14 |
|
15 |
You can use it like this:
|
16 |
```python
|