{ "architectures": [ "UltraVBertModel" ], "auto_map": { "AutoConfig": "configuration_ultravbert.UltraVBertConfig", "AutoModel": "modeling_ultravbert.UltraVBertModel" }, "classifier_dropout": 0.0, "classifier_pooling": "cls", "connector_norm": true, "dtype": "bfloat16", "image_token_id": 49191, "initializer_cutoff_factor": 2.0, "initializer_range": 0.02, "model_type": "ultravbert", "pixel_shuffle_factor": 4, "text_config": { "architectures": [ "UltrabertForMaskedLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 0, "classifier_dropout": 0.0, "classifier_pooling": "cls", "dtype": "bfloat16", "eos_token_id": 0, "head_dim": 128, "hidden_act": "silu", "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 1024, "layer_types": [ "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "full_attention" ], "max_position_embeddings": 32768, "mlp_bias": false, "model_type": "ultrabert", "num_attention_heads": 6, "num_hidden_layers": 24, "num_key_value_heads": 6, "pad_token_id": 0, "prediction_head_size": 768, "rms_norm_eps": 1e-06, "rope_parameters": { "full_attention": { "rope_theta": 2560000.0, "rope_type": "default" }, "sliding_attention": { "rope_theta": 10000.0, "rope_type": "default" } }, "sliding_window": 128, "tie_word_embeddings": false, "use_cache": false, "vocab_size": 49280 }, "tie_word_embeddings": false, "transformers_version": "5.21.0.dev0", "vision_config": { "attention_dropout": 0.0, "dtype": "bfloat16", "hidden_act": "gelu_pytorch_tanh", "hidden_size": 768, "image_size": 512, "intermediate_size": 3072, "layer_norm_eps": 1e-06, "model_type": "siglip_vision_model", "num_attention_heads": 12, "num_channels": 3, "num_hidden_layers": 12, "patch_size": 16, "vision_use_head": false }, "vocab_size": 49280 }