Update README.md
Browse files
README.md
CHANGED
|
@@ -28,7 +28,6 @@ This is the preview instruct variant.
|
|
| 28 |
- the model was post trained with an effective batch size of 32, and with a custom home-made ademamix based 2bit fp8 optimizer for real memory savings.
|
| 29 |
- post training was done with a sequence length of 32k, and thus that is the supported max sequence length by the model.
|
| 30 |
## PLB benchmark results:
|
| 31 |
-
|
| 32 |
text:
|
| 33 |
Model|Publisher|Elo
|
| 34 |
|:---|:---:|:---:|
|
|
|
|
| 28 |
- the model was post trained with an effective batch size of 32, and with a custom home-made ademamix based 2bit fp8 optimizer for real memory savings.
|
| 29 |
- post training was done with a sequence length of 32k, and thus that is the supported max sequence length by the model.
|
| 30 |
## PLB benchmark results:
|
|
|
|
| 31 |
text:
|
| 32 |
Model|Publisher|Elo
|
| 33 |
|:---|:---:|:---:|
|