DM05-Lerobot / config.json
Dexmal2026's picture
sync model files
716afe3 verified
Raw
History Blame Contribute Delete
12.7 kB
{
"type": "dm05",
"n_obs_steps": 1,
"input_features": {
"observation.state": {
"type": "STATE",
"shape": [
14
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
14
]
}
},
"device": "cpu",
"use_amp": false,
"use_peft": false,
"push_to_hub": true,
"repo_id": null,
"private": null,
"tags": null,
"license": "gemma",
"pretrained_path": null,
"pretrained_revision": null,
"core_config": {
"vlm_gradient_checkpointing": true,
"ae_gradient_checkpointing": true,
"ae_gradient_checkpointing_layers": 1,
"freeze_vlm_embedding": false,
"llm_attn_implementation": "eager",
"vision_attn_implementation": "sdpa",
"action_attn_implementation": "sdpa",
"transformers_version": "5.5.4",
"architectures": [
"DM05ForConditionalGeneration"
],
"output_hidden_states": false,
"return_dict": true,
"dtype": "bfloat16",
"chunk_size_feed_forward": 0,
"is_encoder_decoder": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"problem_type": null,
"vlm_config": {
"architectures": [
"Gemma3ForConditionalGeneration"
],
"output_hidden_states": false,
"return_dict": true,
"dtype": "bfloat16",
"chunk_size_feed_forward": 0,
"is_encoder_decoder": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"problem_type": null,
"text_config": {
"architectures": null,
"output_hidden_states": false,
"return_dict": true,
"dtype": "bfloat16",
"chunk_size_feed_forward": 0,
"is_encoder_decoder": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"problem_type": null,
"vocab_size": 262208,
"hidden_size": 2560,
"intermediate_size": 10240,
"num_hidden_layers": 34,
"num_attention_heads": 8,
"num_key_value_heads": 4,
"head_dim": 256,
"hidden_activation": "gelu_pytorch_tanh",
"max_position_embeddings": 131072,
"initializer_range": 0.02,
"rms_norm_eps": 1e-06,
"use_cache": true,
"pad_token_id": 0,
"eos_token_id": 1,
"bos_token_id": 2,
"tie_word_embeddings": true,
"rope_parameters": {
"full_attention": {
"factor": 8.0,
"rope_theta": 1000000.0,
"rope_type": "linear"
},
"sliding_attention": {
"rope_theta": 10000.0,
"rope_type": "default"
}
},
"attention_bias": false,
"attention_dropout": 0.0,
"query_pre_attn_scalar": 256,
"sliding_window": 4096,
"layer_types": [
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention"
],
"final_logit_softcapping": null,
"attn_logit_softcapping": null,
"use_bidirectional_attention": false,
"_sliding_window_pattern": 6,
"_name_or_path": "",
"model_type": "gemma3_text",
"attn_implementation": "eager",
"output_attentions": false
},
"vision_config": {
"architectures": null,
"output_hidden_states": false,
"return_dict": true,
"dtype": "bfloat16",
"chunk_size_feed_forward": 0,
"is_encoder_decoder": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"problem_type": null,
"hidden_size": 1152,
"intermediate_size": 4304,
"num_hidden_layers": 27,
"num_attention_heads": 16,
"num_channels": 3,
"image_size": 448,
"patch_size": 14,
"hidden_act": "gelu_pytorch_tanh",
"layer_norm_eps": 1e-06,
"attention_dropout": 0.0,
"_name_or_path": "",
"model_type": "siglip_vision_model",
"vision_use_head": false,
"attn_implementation": "sdpa",
"output_attentions": false
},
"mm_tokens_per_image": 256,
"boi_token_index": 255999,
"eoi_token_index": 256000,
"image_token_index": 262144,
"initializer_range": 0.02,
"tie_word_embeddings": true,
"_name_or_path": "",
"eos_token_id": [
1,
106
],
"model_type": "gemma3",
"attn_implementation": {
"": "eager",
"text_config": "eager",
"vision_config": "sdpa"
},
"output_attentions": false
},
"action_config": {
"architectures": null,
"output_hidden_states": false,
"return_dict": true,
"dtype": null,
"chunk_size_feed_forward": 0,
"is_encoder_decoder": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"problem_type": null,
"vocab_size": 262208,
"hidden_size": 1024,
"intermediate_size": 4096,
"num_hidden_layers": 34,
"num_attention_heads": 8,
"num_key_value_heads": 4,
"head_dim": 256,
"hidden_activation": "gelu_pytorch_tanh",
"max_position_embeddings": 131072,
"initializer_range": 0.02,
"rms_norm_eps": 1e-06,
"use_cache": true,
"pad_token_id": 0,
"eos_token_id": 1,
"bos_token_id": 2,
"tie_word_embeddings": true,
"rope_parameters": {
"full_attention": {
"factor": 8.0,
"rope_theta": 1000000.0,
"rope_type": "linear"
},
"sliding_attention": {
"rope_theta": 10000.0,
"rope_type": "default"
}
},
"attention_bias": false,
"attention_dropout": 0.0,
"query_pre_attn_scalar": 256,
"sliding_window": 4096,
"layer_types": [
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention"
],
"final_logit_softcapping": null,
"attn_logit_softcapping": null,
"use_bidirectional_attention": false,
"_sliding_window_pattern": 6,
"_name_or_path": "",
"model_type": "gemma3_text",
"attn_implementation": "sdpa",
"action_attn_implementation": "sdpa",
"ae_gradient_checkpointing_layers": 1,
"output_attentions": false
},
"processor_config": null,
"action_dim": 32,
"chunk_size": 50,
"bf16": true,
"_name_or_path": ".",
"bos_token_id": 2,
"eos_token_id": 1,
"model_type": "dexbotic_dm05",
"pad_token_id": 0,
"tie_word_embeddings": true,
"gradient_checkpointing": true,
"output_attentions": false
},
"chunk_size": 50,
"n_action_steps": 50,
"drop_n_last_frames": 1,
"max_state_dim": 32,
"max_action_dim": 32,
"image_resolution": [
448,
448
],
"empty_cameras": 0,
"add_state": true,
"pretrained_name_or_path": ".",
"processor_name_or_path": null,
"image_keys": null,
"n_bins": 256,
"tokenizer_max_length": 1024,
"diffusion_steps": 10,
"compile_model": false,
"compile_suffix_pad_length": 1024,
"llm_attn_implementation": "eager",
"vision_attn_implementation": "sdpa",
"action_attn_implementation": "sdpa",
"use_liger_kernel": false,
"dtype": "bfloat16",
"vlm_gradient_checkpointing": true,
"ae_gradient_checkpointing": true,
"ae_gradient_checkpointing_layers": 1,
"freeze_vlm_embedding": true,
"use_relative_actions": false,
"relative_exclude_joints": [
"gripper"
],
"action_feature_names": [
"joint_0",
"joint_1",
"joint_2",
"joint_3",
"joint_4",
"joint_5",
"gripper_6",
"joint_7",
"joint_8",
"joint_9",
"joint_10",
"joint_11",
"joint_12",
"gripper_13"
],
"norm_clip": true,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"STATE": "QUANTILES",
"ACTION": "QUANTILES"
},
"optimizer_lr": 2.5e-05,
"optimizer_betas": [
0.9,
0.95
],
"optimizer_eps": 1e-08,
"optimizer_weight_decay": 1e-10,
"optimizer_grad_clip_norm": 1.0,
"scheduler_warmup_steps": 1000,
"scheduler_decay_steps": 50000,
"scheduler_decay_lr": 2.5e-06
}