| { | |
| "architectures": [ | |
| "Gemma3ForConditionalGeneration" | |
| ], | |
| "boi_token_index": 255999, | |
| "eoi_token_index": 256000, | |
| "image_token_index": 262144, | |
| "initializer_range": 0.02, | |
| "mm_tokens_per_image": 256, | |
| "model_type": "gemma3", | |
| "text_config": { | |
| "head_dim": 32, | |
| "hidden_size": 8, | |
| "intermediate_size": 32, | |
| "model_type": "gemma3_text", | |
| "num_attention_heads": 2, | |
| "num_hidden_layers": 1, | |
| "num_key_value_heads": 2, | |
| "rope_scaling": { | |
| "factor": 8.0, | |
| "rope_type": "linear" | |
| }, | |
| "sliding_window": 1024 | |
| }, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.50.0.dev0", | |
| "vision_config": { | |
| "hidden_size": 16, | |
| "image_size": 896, | |
| "intermediate_size": 32, | |
| "model_type": "siglip_vision_model", | |
| "num_attention_heads": 2, | |
| "num_hidden_layers": 1, | |
| "patch_size": 14 | |
| } | |
| } | |