|
{ |
|
"_name_or_path": "ckpts/clip-cifar10-1/accuracy_best_model/", |
|
"architectures": [ |
|
"CLIPModel" |
|
], |
|
"elastic_config": { |
|
"text": { |
|
"atten_out_space": [ |
|
"None" |
|
], |
|
"inter_hidden_space": [ |
|
2048, |
|
1536, |
|
1024 |
|
], |
|
"residual_hidden_space": [ |
|
"None" |
|
] |
|
}, |
|
"vision": { |
|
"atten_out_space": [ |
|
"None" |
|
], |
|
"inter_hidden_space": [ |
|
3072, |
|
2048, |
|
1280 |
|
], |
|
"residual_hidden_space": [ |
|
"None" |
|
] |
|
} |
|
}, |
|
"initializer_factor": 1.0, |
|
"logit_scale_init_value": 2.6592, |
|
"model_type": "clip", |
|
"projection_dim": 512, |
|
"text_config": { |
|
"bos_token_id": 0, |
|
"dropout": 0.0, |
|
"eos_token_id": 2, |
|
"model_type": "clip_text_model" |
|
}, |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.40.0.dev0", |
|
"vision_config": { |
|
"dropout": 0.0, |
|
"model_type": "clip_vision_model" |
|
} |
|
} |
|
|