Update README.md
#6
by
MaziyarPanahi
- opened
README.md
CHANGED
|
@@ -30,6 +30,21 @@ datasets:
|
|
| 30 |
|
| 31 |
This model is a fine-tune (DPO) of `meta-llama/Meta-Llama-3-8B-Instruct` model.
|
| 32 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
# How to use
|
| 34 |
|
| 35 |
You can use this model by using `MaziyarPanahi/Llama-3-8B-Instruct-DPO-v0.1` as the model name in Hugging Face's
|
|
|
|
| 30 |
|
| 31 |
This model is a fine-tune (DPO) of `meta-llama/Meta-Llama-3-8B-Instruct` model.
|
| 32 |
|
| 33 |
+
# Prompt Template
|
| 34 |
+
|
| 35 |
+
This model uses `ChatML` prompt template:
|
| 36 |
+
|
| 37 |
+
```
|
| 38 |
+
<|im_start|>system
|
| 39 |
+
{System}
|
| 40 |
+
<|im_end|>
|
| 41 |
+
<|im_start|>user
|
| 42 |
+
{User}
|
| 43 |
+
<|im_end|>
|
| 44 |
+
<|im_start|>assistant
|
| 45 |
+
{Assistant}
|
| 46 |
+
````
|
| 47 |
+
|
| 48 |
# How to use
|
| 49 |
|
| 50 |
You can use this model by using `MaziyarPanahi/Llama-3-8B-Instruct-DPO-v0.1` as the model name in Hugging Face's
|