{ "action_dim": 32, "chunk_size": 50, "image_aspect_ratio": "pad", "mm_hidden_size": 1152, "mm_projector_lr": null, "mm_projector_type": "linear", "model_type": "dexbotic_pi0", "tokenizer_model_max_length": 2048, "processor_config": "google/siglip-so400m-patch14-224", "tokenizer_padding_side": "right", "tune_mm_mlp_adapter": false, "unfreeze_vision_tower": false, "use_cache": false, "use_mm_proj": true, "vm_loss_enable": false, "assistant_token_number": -1, "llm_config": { "model_type": "gemma", "bos_token_id": 2, "eos_token_id": 1, "attention_dropout": 0.0, "attention_bias": false, "head_dim": 256, "image_token_len": 256, "initializer_range": 0.02, "intermediate_size": 16384, "max_position_embeddings": 8192, "hidden_act": "gelu_pytorch_tanh", "hidden_activation": null, "hidden_size": 2048, "num_attention_heads": 8, "num_hidden_layers": 18, "num_key_value_heads": 1, "pad_token_id": 0, "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 10000.0, "torch_dtype": "float32", "vocab_size": 257152 }, "vision_config": { "vision_use_head": false, "hidden_size": 1152, "intermediate_size": 4304, "model_type": "siglip_vision_model", "hidden_act": "gelu_fast", "num_attention_heads": 16, "num_hidden_layers": 27, "patch_size": 14 }, "action_config": { "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 2, "eos_token_id": 1, "head_dim": 256, "hidden_act": "gelu_pytorch_tanh", "hidden_size": 1024, "initializer_range": 0.02, "intermediate_size": 4096, "max_position_embeddings": 8192, "model_type": "gemma", "num_attention_heads": 8, "num_hidden_layers": 18, "num_key_value_heads": 1, "pad_token_id": 0, "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 10000.0, "torch_dtype": "float32", "transformers_version": "4.38.0.dev0", "use_cache": true, "vocab_size": 256000 } }