Update README.md
Browse files
README.md
CHANGED
@@ -183,8 +183,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
183 |
|
184 |
# Model Card for Zephyr 7B β
|
185 |
|
|
|
|
|
186 |
Zephyr is a series of language models that are trained to act as helpful assistants. Zephyr-7B-β is the second model in the series, and is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) that was trained on on a mix of publicly available, synthetic datasets using [Direct Preference Optimization (DPO)](https://arxiv.org/abs/2305.18290). We found that removing the in-built alignment of these datasets boosted performance on [MT Bench](https://huggingface.co/spaces/lmsys/mt-bench) and made the model more helpful. However, this means that model is likely to generate problematic text when prompted to do so. You can find more details in the [technical report](https://arxiv.org/abs/2310.16944).
|
187 |
-
|
188 |
|
189 |
## Model description
|
190 |
|
|
|
183 |
|
184 |
# Model Card for Zephyr 7B β
|
185 |
|
186 |
+
**This repo variant are quantization experiments based on the original model.**
|
187 |
+
|
188 |
Zephyr is a series of language models that are trained to act as helpful assistants. Zephyr-7B-β is the second model in the series, and is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) that was trained on on a mix of publicly available, synthetic datasets using [Direct Preference Optimization (DPO)](https://arxiv.org/abs/2305.18290). We found that removing the in-built alignment of these datasets boosted performance on [MT Bench](https://huggingface.co/spaces/lmsys/mt-bench) and made the model more helpful. However, this means that model is likely to generate problematic text when prompted to do so. You can find more details in the [technical report](https://arxiv.org/abs/2310.16944).
|
189 |
+
|
190 |
|
191 |
## Model description
|
192 |
|