HARP_VLA_calvin / config.json
ypz21's picture
Release merged HARP-VLA CALVIN inference checkpoint
a323c0a verified
Raw History Blame Contribute Delete
1.9 kB
{
"arch_specifier": "no-align+fused-gelu-mlp",
"architectures": [
"OpenVLAForActionPrediction"
],
"hf_llm_id": "meta-llama/Llama-2-7b-hf",
"image_resize_strategy": "resize-naive",
"image_sizes": [
224,
224
],
"llm_backbone_id": "llama2-7b-pure",
"llm_max_length": 2048,
"model_type": "openvla",
"n_action_bins": 256,
"norm_stats": {
"calvin": {
"action": {
"q01": [
-0.7241964542398116,
-0.5942806773882542,
-0.4188806280352448,
-0.42482009663749476,
-0.45296629479497935,
-0.9627345511434341,
-1.0
],
"q99": [
0.7266286401209516,
0.5572942865369028,
0.5975622582326067,
0.41687738248912865,
0.3928198217740266,
0.9829434784501938,
1.0
]
},
"proprio": {
"q01": [
-0.28179742900457466,
-0.4093728881099912,
0.33205914462542957,
-3.1394471604896506,
-0.29085095349099827,
0.3954914036933885,
-1.0
],
"q99": [
0.29786006625497197,
0.07310697205401105,
0.6392391304035469,
3.1395528800861103,
0.19396112494725273,
2.818498526690071,
1.0
]
}
}
},
"output_projector_states": false,
"pad_to_multiple_of": 64,
"pad_token_id": 32000,
"text_config": {
"model_type": "llama",
"pad_token_id": 32000,
"torch_dtype": "bfloat16",
"vocab_size": 32064
},
"timm_model_ids": [
"vit_large_patch14_reg4_dinov2.lvd142m",
"vit_so400m_patch14_siglip_224"
],
"timm_override_act_layers": [
null,
null
],
"torch_dtype": "bfloat16",
"transformers_version": "4.40.1",
"use_fused_vision_backbone": true,
"vision_backbone_id": "dinosiglip-vit-so-224px"
}