| { |
| "architectures": [ |
| "FineViTModel" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_finevit.FineViTConfig", |
| "AutoModel": "modeling_finevit.FineViTModel" |
| }, |
| "backbone_name": "dinov2_base_reg", |
| "backbone_patch_size": 14, |
| "backbone_pool_size": 2, |
| "backbone_pretrained": true, |
| "decoder_head_dim": 64, |
| "decoder_num_attention_heads": 12, |
| "decoder_num_layers": 8, |
| "decoder_position_grid_size": 16, |
| "dropout": 0.0, |
| "dtype": "float32", |
| "encoder_head_dim": 64, |
| "encoder_num_attention_heads": 12, |
| "encoder_num_layers": 8, |
| "encoder_position_grid_size": 16, |
| "intermediate_size": 3072, |
| "model_type": "finevit", |
| "prefix_alignment_mlp_ratio": 4.0, |
| "prefix_alignment_num_attention_heads": 12, |
| "prefix_alignment_qk_norm": true, |
| "training_mask_ratio": 0.0, |
| "training_noise_gamma": 0.0, |
| "transformers_version": "5.8.1" |
| } |
|
|
