cross link to meg-ds checkpoint
Browse files
README.md
CHANGED
|
@@ -2442,6 +2442,11 @@ Initial prompting experiments using interim checkpoints: https://huggingface.co/
|
|
| 2442 |
|
| 2443 |
</details>
|
| 2444 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2445 |
---
|
| 2446 |
|
| 2447 |
# Model Card Authors
|
|
|
|
| 2442 |
|
| 2443 |
</details>
|
| 2444 |
|
| 2445 |
+
|
| 2446 |
+
## Original checkpoints
|
| 2447 |
+
|
| 2448 |
+
The checkpoints in this repo correspond to the HuggingFace Transformers format. If you want to use our fork of [Megatron-DeepSpeed](https://github.com/bigscience-workshop/Megatron-DeepSpeed) that the model was trained with, you'd want to use [this repo instead](https://huggingface.co/bigscience/bloom-optimizer-states).
|
| 2449 |
+
|
| 2450 |
---
|
| 2451 |
|
| 2452 |
# Model Card Authors
|