jwieczorekhabana
commited on
Commit
•
d0b8563
1
Parent(s):
22ac3ad
Update README.md
Browse files
README.md
CHANGED
@@ -15,8 +15,7 @@ This model only contains the `GaudiConfig` file for running the [distilbert-base
|
|
15 |
This enables to specify:
|
16 |
- `use_fused_adam`: whether to use Habana's custom AdamW implementation
|
17 |
- `use_fused_clip_norm`: whether to use Habana's fused gradient norm clipping operator
|
18 |
-
- `
|
19 |
-
In those cases this parameter is already present in huggingface topology Habana gaudi_config.json.
|
20 |
|
21 |
## Usage
|
22 |
|
|
|
15 |
This enables to specify:
|
16 |
- `use_fused_adam`: whether to use Habana's custom AdamW implementation
|
17 |
- `use_fused_clip_norm`: whether to use Habana's fused gradient norm clipping operator
|
18 |
+
- `use_torch_autocast`: whether to use Torch Autocast for managing mixed precision
|
|
|
19 |
|
20 |
## Usage
|
21 |
|