Update README.md
Browse files
README.md
CHANGED
@@ -143,9 +143,6 @@ For instruction training, we first trained the model with Supervised Fine-tuning
|
|
143 |
</table>
|
144 |
|
145 |
## Interfacing with the Instruct Model
|
146 |
-
Model weights were converted from the original Mamba2 implementation to be Hugging Face compatible. <br>
|
147 |
-
Due to the lack of official support for Mamba2 attention layers in Hugging Face Transformers, custom modeling files are included. <br>
|
148 |
-
The attention layer implementation for the modeling files are based on the work from Pull Request #32027 in the Hugging Face Transformers repository: [https://github.com/huggingface/transformers/pull/32027](https://github.com/huggingface/transformers/pull/32027)
|
149 |
|
150 |
> [!IMPORTANT]
|
151 |
> To ensure optimal performance, please use the following template when interacting with the model:
|
@@ -199,7 +196,9 @@ The play "Romeo and Juliet" by William Shakespeare is a classic example of a tra
|
|
199 |
```
|
200 |
|
201 |
## Usage with HuggingFace transformers
|
202 |
-
Model weights were converted
|
|
|
|
|
203 |
|
204 |
To speed up inference, we recommend installing mamba-ssm and flash attention 2.
|
205 |
|
|
|
143 |
</table>
|
144 |
|
145 |
## Interfacing with the Instruct Model
|
|
|
|
|
|
|
146 |
|
147 |
> [!IMPORTANT]
|
148 |
> To ensure optimal performance, please use the following template when interacting with the model:
|
|
|
196 |
```
|
197 |
|
198 |
## Usage with HuggingFace transformers
|
199 |
+
Model weights were converted from the original Mamba2 implementation to be Hugging Face compatible. <br>
|
200 |
+
Due to the lack of official support for Mamba2 attention layers in Hugging Face Transformers, custom modeling files are included. <br>
|
201 |
+
The attention layer implementation for the modeling files are based on the work from Pull Request #32027 in the Hugging Face Transformers repository: [https://github.com/huggingface/transformers/pull/32027](https://github.com/huggingface/transformers/pull/32027)
|
202 |
|
203 |
To speed up inference, we recommend installing mamba-ssm and flash attention 2.
|
204 |
|