Upload folder using huggingface_hub
Browse files
README.md
CHANGED
|
@@ -7,13 +7,10 @@ widget:
|
|
| 7 |
example_title: Hello world
|
| 8 |
group: Python
|
| 9 |
base_model:
|
| 10 |
-
- deepseek-ai/DeepSeek-V3.1
|
| 11 |
---
|
| 12 |
|
| 13 |
-
This tiny model is for debugging. It is randomly initialized with the config adapted from [deepseek-ai/DeepSeek-V3.1
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
⚠: Under construction. The current state is not fully verified.
|
| 17 |
|
| 18 |
### Example usage:
|
| 19 |
|
|
@@ -65,7 +62,7 @@ from transformers import (
|
|
| 65 |
set_seed,
|
| 66 |
)
|
| 67 |
from transformers.models.glm4_moe.modeling_glm4_moe import Glm4MoeRMSNorm
|
| 68 |
-
source_model_id = "deepseek-ai/DeepSeek-V3.1
|
| 69 |
save_folder = "/tmp/yujiepan/deepseek-v3.1-tiny-random"
|
| 70 |
|
| 71 |
Path(save_folder).mkdir(parents=True, exist_ok=True)
|
|
|
|
| 7 |
example_title: Hello world
|
| 8 |
group: Python
|
| 9 |
base_model:
|
| 10 |
+
- deepseek-ai/DeepSeek-V3.1
|
| 11 |
---
|
| 12 |
|
| 13 |
+
This tiny model is for debugging. It is randomly initialized with the config adapted from [deepseek-ai/DeepSeek-V3.1](https://huggingface.co/deepseek-ai/DeepSeek-V3.1).
|
|
|
|
|
|
|
|
|
|
| 14 |
|
| 15 |
### Example usage:
|
| 16 |
|
|
|
|
| 62 |
set_seed,
|
| 63 |
)
|
| 64 |
from transformers.models.glm4_moe.modeling_glm4_moe import Glm4MoeRMSNorm
|
| 65 |
+
source_model_id = "deepseek-ai/DeepSeek-V3.1"
|
| 66 |
save_folder = "/tmp/yujiepan/deepseek-v3.1-tiny-random"
|
| 67 |
|
| 68 |
Path(save_folder).mkdir(parents=True, exist_ok=True)
|