{ "type": "rewact", "n_obs_steps": 1, "normalization_mapping": { "VISUAL": "MEAN_STD", "STATE": "MEAN_STD", "ACTION": "MEAN_STD" }, "input_features": { "observation.state": { "type": "STATE", "shape": [ 6 ] }, "observation.images.top": { "type": "VISUAL", "shape": [ 3, 480, 640 ] }, "observation.images.wrist": { "type": "VISUAL", "shape": [ 3, 480, 640 ] } }, "output_features": { "action": { "type": "ACTION", "shape": [ 6 ] } }, "device": null, "use_amp": false, "push_to_hub": true, "repo_id": "pravsels/so100_rewact_sam3", "private": null, "tags": null, "license": null, "chunk_size": 100, "n_action_steps": 100, "vision_backbone": "sam3_vit_l", "pretrained_backbone_weights": "./sam3.pt", "replace_final_stride_with_dilation": false, "pre_norm": false, "dim_model": 512, "n_heads": 8, "dim_feedforward": 3200, "feedforward_activation": "relu", "n_encoder_layers": 4, "n_decoder_layers": 1, "use_vae": true, "latent_dim": 32, "n_vae_encoder_layers": 4, "temporal_ensemble_coeff": null, "dropout": 0.1, "kl_weight": 10.0, "optimizer_lr": 5e-06, "optimizer_weight_decay": 0.0001, "optimizer_lr_backbone": 5e-06, "use_reward_head": true, "reward_loss_weight": 2.0, "vision_encoder_type": "sam3", "freeze_vision_encoder": true, "dinov3": null, "vjepa2": null, "sam3": { "variant": "vit_l", "weights": "./sam3.pt", "fpn_level": 2, "pos_embed_type": "act_sinusoidal", "patch_merge_stages": 2, "use_patch_merge": false, "input_resolution": 1008, "compile_backbone": false } }