Merge branch 'main' of https://huggingface.co/j879/StableBeluga2-bf16
Browse files
README.md
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- conceptofmind/cot_submix_original
|
| 4 |
+
- conceptofmind/flan2021_submix_original
|
| 5 |
+
- conceptofmind/t0_submix_original
|
| 6 |
+
- conceptofmind/niv2_submix_original
|
| 7 |
+
---
|
| 8 |
+
language:
|
| 9 |
+
- en
|
| 10 |
+
pipeline_tag: text-generation
|
| 11 |
+
---
|
| 12 |
+
# Stable Beluga 2 bf16
|
| 13 |
+
|
| 14 |
+
This is bfloat16 safetensor format weight of original stabilityai/StableBeluga2
|
| 15 |
+
|
| 16 |
+
The weight has been sliced to a smaller shard with 1GB each.
|
| 17 |
+
|
| 18 |
+
Total size is half of original float32 weight for faster loading.
|
| 19 |
+
|
| 20 |
+
Please refer to the original model author https://huggingface.co/stabilityai/StableBeluga2
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
```bibtex
|
| 24 |
+
@misc{StableBelugaModels,
|
| 25 |
+
url={[https://huggingface.co/stabilityai/StableBeluga2](https://huggingface.co/stabilityai/StableBeluga2)},
|
| 26 |
+
title={Stable Beluga models},
|
| 27 |
+
author={Mahan, Dakota and Carlow, Ryan and Castricato, Louis and Cooper, Nathan and Laforte, Christian}
|
| 28 |
+
}
|
| 29 |
+
```
|