Update README.md
Browse files
README.md
CHANGED
|
@@ -42,9 +42,14 @@ model = AutoModelForCausalLM.from_pretrained("sambanovasystems/SambaLingo-Turkis
|
|
| 42 |
- Repetition penalty: 1.0
|
| 43 |
- Top-p: 0.9
|
| 44 |
|
| 45 |
-
###
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
|
| 47 |
-
## Evaluation Results
|
| 48 |
|
| 49 |
## Training Details
|
| 50 |
The alignment phase follows the recipe for [Zephyr-7B](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta), and comprises two stages: supervised fine-tuning (SFT) and Direct Performance Optimization (DPO).
|
|
|
|
| 42 |
- Repetition penalty: 1.0
|
| 43 |
- Top-p: 0.9
|
| 44 |
|
| 45 |
+
### Prompting Guideline
|
| 46 |
+
To prompt this model, please use the chat template
|
| 47 |
+
```
|
| 48 |
+
<|user>:\n{question}</s>\n<|assistant|>\n
|
| 49 |
+
```
|
| 50 |
+
|
| 51 |
+
### Example Prompts and Generations
|
| 52 |
|
|
|
|
| 53 |
|
| 54 |
## Training Details
|
| 55 |
The alignment phase follows the recipe for [Zephyr-7B](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta), and comprises two stages: supervised fine-tuning (SFT) and Direct Performance Optimization (DPO).
|