| { |
| "action_dim": 128, |
| "action_head_cfg": { |
| "action_dim": 128, |
| "action_horizon": 16, |
| "add_pos_embed": true, |
| "backbone_embedding_dim": 2048, |
| "correct_vl_mask": false, |
| "diffusion_model_cfg": { |
| "attention_head_dim": 48, |
| "cross_attention_dim": 2048, |
| "dropout": 0.2, |
| "final_dropout": true, |
| "interleave_self_attention": true, |
| "norm_type": "ada_norm", |
| "num_attention_heads": 32, |
| "num_layers": 16, |
| "output_dim": 1024, |
| "positional_embeddings": null |
| }, |
| "hidden_size": 1024, |
| "input_embedding_dim": 1536, |
| "max_action_dim": 128, |
| "max_state_dim": 128, |
| "model_dtype": "float32", |
| "noise_beta_alpha": 1.5, |
| "noise_beta_beta": 1.0, |
| "noise_s": 0.999, |
| "num_inference_timesteps": 4, |
| "num_target_vision_tokens": 0, |
| "num_timestep_buckets": 1000, |
| "tune_diffusion_model": true, |
| "tune_projector": true, |
| "use_vl_mask": false, |
| "use_vlln": true, |
| "vl_self_attention_cfg": { |
| "attention_head_dim": 64, |
| "dropout": 0.2, |
| "final_dropout": true, |
| "num_attention_heads": 32, |
| "num_layers": 4, |
| "positional_embeddings": null |
| } |
| }, |
| "action_horizon": 16, |
| "architectures": [ |
| "GR00T_N1_5_DIAL" |
| ], |
| "attn_implementation": null, |
| "backbone_cfg": { |
| "load_bf16": false, |
| "project_to_dim": null, |
| "reproject_vision": false, |
| "select_layer": 36, |
| "tune_all_llm_embedding": false, |
| "tune_bridge_embedding": true, |
| "tune_llm": true, |
| "tune_visual": false, |
| "use_flash_attention": true, |
| "vlm_path": "Qwen/Qwen2.5-VL-3B-Instruct" |
| }, |
| "bridge_cfg": { |
| "action_only_one_obs": false, |
| "bridge_loss_decay_steps": null, |
| "bridge_loss_end_w": null, |
| "bridge_loss_type": "mse", |
| "bridge_type": "end2end", |
| "compute_bridge_loss": true, |
| "goal_image_type": "future", |
| "noise_tau": 0, |
| "num_bridge_tokens": 64, |
| "omit_image_type_embedding_for_goal": false, |
| "reweight_noise": false, |
| "tokenizer_len": 151729, |
| "tune_bridge_goal": false, |
| "tune_bridge_visual": false, |
| "tune_image_type_embedding": true, |
| "unified_embodiment_id": null, |
| "use_bridge": true, |
| "use_image_type_embedding": true, |
| "use_separate_projector_for_loss": true |
| }, |
| "compute_dtype": "bfloat16", |
| "hidden_size": 2048, |
| "ignore_lang_prefix": true, |
| "model_dtype": "float32", |
| "model_type": "gr00t_n1_5_dial", |
| "torch_dtype": "float32", |
| "transformers_version": "4.52.0", |
| "video_delta_indices": [ |
| 0 |
| ] |
| } |
