{ "feature_type": "latent_student_spatial_kv", "conditioning_variant": "b3", "config": "configs/b2_rdt1b_lora.yaml", "root": "dataset/datasets", "splits": [ "train", "validation", "test" ], "datasets": [ "libero_spatial" ], "seed": 42, "stage": null, "normalize_actions": false, "action_target_mode": "delta", "state_dim": 9, "action_dim": 7, "proprioception_schema": "libero_joint7_gripper2_norm01_action7_v1", "state_encoder_layout": "libero_ortho6d", "action_encoder_layout": "libero_ortho6d", "gripper_processing": "two finger qpos normalized independently with shared travel min/max to [0,1]; raw HDF5 action command", "max_samples_per_episode": null, "all_samples_per_episode": true, "gripper_window_before": 3, "gripper_window_after": 3, "gripper_change_scope": "directional", "open_to_close_before": 10, "open_to_close_after": 11, "close_to_open_before": 10, "close_to_open_after": 11, "student_model_id": "/workspace/model/Latent-Student-Spatial-Forcing-fixed/", "processor_id": "/workspace/model/stage1_unsloth", "spatial_parameters_path": null, "latent_count": 6, "spatial_token_count": 5, "layer_index": 7, "prompt_template": "You are a robot manipulation assistant. Given an observation image and a task instruction, predict the end-effector's 2D trajectory as 5 waypoints. Output ONLY the coordinate list in this exact format: [[x1,y1],[x2,y2],[x3,y3],[x4,y4],[x5,y5]]\n\nTask: The task is {task}. What is the trajectory that the end effector should take?", "qwen_kv_dim": 2048, "qwen_hidden_state_dim": 2560, "waypoint_dim": 2, "has_qwen_hidden_states": true, "has_latent_waypoints": true, "include_t5": true, "t5_precision": "bf16", "t5_batch_size": 32, "cache_image_slots": true, "image_history_size": 2, "max_images_per_sample": 6, "image_storage_codec": "raw_uint8", "image_storage_lossless": true, "image_storage_compressed": false, "image_jpeg_quality": null, "cache_layout": "sample_shard", "batch_size": 32 }