Instructions to use FashGate/config with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FashGate/config with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("FashGate/config", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update config.json
Browse files- config.json +1 -1
config.json
CHANGED
|
@@ -91,7 +91,7 @@
|
|
| 91 |
"randn_source": "GPU",
|
| 92 |
"cross_attention_optimization": "Automatic",
|
| 93 |
"s_min_uncond": 0.0,
|
| 94 |
-
"token_merging_ratio": 0.
|
| 95 |
"token_merging_ratio_img2img": 0.0,
|
| 96 |
"token_merging_ratio_hr": 0.0,
|
| 97 |
"pad_cond_uncond": false,
|
|
|
|
| 91 |
"randn_source": "GPU",
|
| 92 |
"cross_attention_optimization": "Automatic",
|
| 93 |
"s_min_uncond": 0.0,
|
| 94 |
+
"token_merging_ratio": 0.5,
|
| 95 |
"token_merging_ratio_img2img": 0.0,
|
| 96 |
"token_merging_ratio_hr": 0.0,
|
| 97 |
"pad_cond_uncond": false,
|