{
    "type": "gr00t_n1d6",
    "n_obs_steps": 1,
    "input_features": {
        "observation.state": {
            "type": "STATE",
            "shape": [
                6
            ]
        },
        "observation.images.front": {
            "type": "VISUAL",
            "shape": [
                3,
                480,
                640
            ]
        },
        "observation.images.wrist": {
            "type": "VISUAL",
            "shape": [
                3,
                480,
                640
            ]
        }
    },
    "output_features": {
        "action": {
            "type": "ACTION",
            "shape": [
                6
            ]
        }
    },
    "device": "cuda",
    "use_amp": false,
    "use_peft": false,
    "push_to_hub": false,
    "repo_id": null,
    "private": null,
    "tags": null,
    "license": null,
    "pretrained_path": null,
    "chunk_size": 40,
    "n_action_steps": 40,
    "max_state_dim": 29,
    "max_action_dim": 29,
    "normalization_mapping": {
        "VISUAL": "IDENTITY",
        "STATE": "MEAN_STD",
        "ACTION": "MEAN_STD"
    },
    "image_size": [
        224,
        224
    ],
    "base_model_path": "nvidia/GR00T-N1.6-3B",
    "tokenizer_assets_repo": "nvidia/Eagle-Block2A-2B-v2",
    "backbone_model_type": "eagle",
    "model_revision": null,
    "backbone_embedding_dim": 2048,
    "select_layer": 16,
    "reproject_vision": false,
    "use_flash_attention": true,
    "load_bf16": true,
    "eagle_collator": false,
    "backbone_trainable_params_fp32": true,
    "embodiment_tag": "new_embodiment",
    "tune_top_llm_layers": 4,
    "tune_llm": false,
    "tune_visual": false,
    "tune_projector": true,
    "tune_diffusion_model": true,
    "tune_vlln": true,
    "image_crop_size": null,
    "image_target_size": null,
    "shortest_image_edge": 256,
    "crop_fraction": 0.95,
    "random_rotation_angle": null,
    "color_jitter_params": null,
    "use_albumentations_transforms": true,
    "formalize_language": true,
    "lora_rank": 0,
    "lora_alpha": 16,
    "lora_dropout": 0.1,
    "lora_full_model": false,
    "action_horizon": 16,
    "hidden_size": 1024,
    "input_embedding_dim": 1536,
    "add_pos_embed": true,
    "attn_dropout": 0.2,
    "use_vlln": true,
    "max_seq_len": 1024,
    "use_alternate_vl_dit": true,
    "attend_text_every_n_blocks": 2,
    "diffusion_model_cfg": {
        "positional_embeddings": null,
        "num_layers": 32,
        "num_attention_heads": 32,
        "attention_head_dim": 48,
        "norm_type": "ada_norm",
        "dropout": 0.2,
        "final_dropout": true,
        "output_dim": 1024,
        "interleave_self_attention": true
    },
    "num_inference_timesteps": 4,
    "noise_beta_alpha": 1.5,
    "noise_beta_beta": 1.0,
    "noise_s": 0.999,
    "num_timestep_buckets": 1000,
    "state_dropout_prob": 0.0,
    "state_additive_noise_scale": 0.0,
    "apply_sincos_state_encoding": false,
    "use_relative_action": true,
    "max_num_embodiments": 32,
    "optimizer_lr": 0.0001,
    "optimizer_betas": [
        0.95,
        0.999
    ],
    "optimizer_eps": 1e-08,
    "optimizer_weight_decay": 1e-05,
    "warmup_ratio": 0.05,
    "use_bf16": true,
    "video_backend": "decord",
    "balance_dataset_weights": true,
    "balance_trajectory_weights": true,
    "dataset_paths": null,
    "output_dir": "./tmp/gr00t_n16",
    "save_steps": 1000,
    "max_steps": 10000,
    "batch_size": 32,
    "dataloader_num_workers": 8,
    "report_to": "wandb",
    "resume": false
}