Robotics
LeRobot
Safetensors
nanovlm_value_function
reward-model
lilkm's picture
lilkm HF Staff
Upload reward model weights, train config and readme
c52aff8 verified
Raw
History Blame Contribute Delete
1.33 kB
{
"type": "nanovlm_value_function",
"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": {},
"device": "cuda",
"pretrained_path": null,
"pretrained_revision": null,
"push_to_hub": true,
"repo_id": "lilkm/vf_stackblocks_nanovlm_probe_hl_v2",
"license": null,
"tags": null,
"private": null,
"nanovlm_pretrained_path": "lusxvr/nanoVLM-460M-8k",
"nanovlm_code_path": "third_party/nanoVLM",
"tokenizer_max_length": 8192,
"num_value_bins": 201,
"value_support_min": -1.0,
"value_support_max": 0.0,
"hl_gauss_sigma_ratio": 0.75,
"target_method": "hl_gauss",
"use_one_hot_terminal": false,
"value_dropout": 0.0,
"freeze_vision_encoder": true,
"freeze_multimodal_projector": true,
"freeze_language_model": true,
"normalization_mapping": {
"VISUAL": "IDENTITY"
}
}