Zero-Shot Image Classification
Transformers
Safetensors
clip
Inference Endpoints
CLIP-SAE-ViT-L-14 / config.json
zer0int's picture
SAE-informed fine-tune
114b1ef verified
raw
history blame contribute delete
736 Bytes
{
"_name_or_path": "CLIP-SAE-ViT-L-14/",
"architectures": [
"CLIPModel"
],
"initializer_factor": 1.0,
"logit_scale_init_value": 4.6052,
"model_type": "clip",
"projection_dim": 768,
"text_config": {
"dropout": 0.0,
"hidden_size": 768,
"intermediate_size": 3072,
"model_type": "clip_text_model",
"num_attention_heads": 12,
"projection_dim": 768
},
"torch_dtype": "float32",
"transformers_version": "4.46.3",
"vision_config": {
"dropout": 0.0,
"hidden_size": 1024,
"intermediate_size": 4096,
"model_type": "clip_vision_model",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"patch_size": 14,
"projection_dim": 768
}
}