config.json
| 1 | { |
| 2 | "_name_or_path": "openvla/openvla-7b", |
| 3 | "arch_specifier": "no-align+fused-gelu-mlp", |
| 4 | "architectures": [ |
| 5 | "OpenVLAForActionPrediction" |
| 6 | ], |
| 7 | "auto_map": { |
| 8 | "AutoConfig": "openvla/openvla-7b--configuration_prismatic.OpenVLAConfig", |
| 9 | "AutoModelForVision2Seq": "openvla/openvla-7b--modeling_prismatic.OpenVLAForActionPrediction" |
| 10 | }, |
| 11 | "hf_llm_id": "meta-llama/Llama-2-7b-hf", |
| 12 | "image_resize_strategy": "resize-naive", |
| 13 | "image_sizes": [ |
| 14 | 224, |
| 15 | 224 |
| 16 | ], |
| 17 | "llm_backbone_id": "llama2-7b-pure", |
| 18 | "llm_max_length": 2048, |
| 19 | "model_type": "openvla", |
| 20 | "n_action_bins": 256, |
| 21 | "norm_stats": { |
| 22 | "libero_object": { |
| 23 | "action": { |
| 24 | "mean": [ |
| 25 | 0.07096529006958008, |
| 26 | 0.13498851656913757, |
| 27 | -0.04601382836699486, |
| 28 | 0.00123520044144243, |
| 29 | 0.006998839322477579, |
| 30 | -0.015027612447738647, |
| 31 | 0.46428999304771423 |
| 32 | ], |
| 33 | "std": [ |
| 34 | 0.2681235373020172, |
| 35 | 0.43846824765205383, |
| 36 | 0.4474974274635315, |
| 37 | 0.024446550756692886, |
| 38 | 0.049355510622262955, |
| 39 | 0.042107198387384415, |
| 40 | 0.49879148602485657 |
| 41 | ], |
| 42 | "max": [ |
| 43 | 0.9375, |
| 44 | 0.8919642567634583, |
| 45 | 0.9375, |
| 46 | 0.17678570747375488, |
| 47 | 0.35035714507102966, |
| 48 | 0.1810714304447174, |
| 49 | 1.0 |
| 50 | ], |
| 51 | "min": [ |
| 52 | -0.8839285969734192, |
| 53 | -0.9375, |
| 54 | -0.9375, |
| 55 | -0.15000000596046448, |
| 56 | -0.29035714268684387, |
| 57 | -0.32892856001853943, |
| 58 | 0.0 |
| 59 | ], |
| 60 | "q01": [ |
| 61 | -0.5383928418159485, |
| 62 | -0.8758928775787354, |
| 63 | -0.9375, |
| 64 | -0.06964285671710968, |
| 65 | -0.11678571254014969, |
| 66 | -0.15964286029338837, |
| 67 | 0.0 |
| 68 | ], |
| 69 | "q99": [ |
| 70 | 0.8464285731315613, |
| 71 | 0.84375, |
| 72 | 0.9375, |
| 73 | 0.08142857253551483, |
| 74 | 0.14892856776714325, |
| 75 | 0.0867857113480568, |
| 76 | 1.0 |
| 77 | ], |
| 78 | "mask": [ |
| 79 | true, |
| 80 | true, |
| 81 | true, |
| 82 | true, |
| 83 | true, |
| 84 | true, |
| 85 | false |
| 86 | ] |
| 87 | }, |
| 88 | "proprio": { |
| 89 | "mean": [ |
| 90 | 0.0, |
| 91 | 0.0, |
| 92 | 0.0, |
| 93 | 0.0, |
| 94 | 0.0, |
| 95 | 0.0, |
| 96 | 0.0 |
| 97 | ], |
| 98 | "std": [ |
| 99 | 0.0, |
| 100 | 0.0, |
| 101 | 0.0, |
| 102 | 0.0, |
| 103 | 0.0, |
| 104 | 0.0, |
| 105 | 0.0 |
| 106 | ], |
| 107 | "max": [ |
| 108 | 0.0, |
| 109 | 0.0, |
| 110 | 0.0, |
| 111 | 0.0, |
| 112 | 0.0, |
| 113 | 0.0, |
| 114 | 0.0 |
| 115 | ], |
| 116 | "min": [ |
| 117 | 0.0, |
| 118 | 0.0, |
| 119 | 0.0, |
| 120 | 0.0, |
| 121 | 0.0, |
| 122 | 0.0, |
| 123 | 0.0 |
| 124 | ], |
| 125 | "q01": [ |
| 126 | 0.0, |
| 127 | 0.0, |
| 128 | 0.0, |
| 129 | 0.0, |
| 130 | 0.0, |
| 131 | 0.0, |
| 132 | 0.0 |
| 133 | ], |
| 134 | "q99": [ |
| 135 | 0.0, |
| 136 | 0.0, |
| 137 | 0.0, |
| 138 | 0.0, |
| 139 | 0.0, |
| 140 | 0.0, |
| 141 | 0.0 |
| 142 | ] |
| 143 | }, |
| 144 | "num_transitions": 66984, |
| 145 | "num_trajectories": 454 |
| 146 | } |
| 147 | }, |
| 148 | "output_projector_states": false, |
| 149 | "pad_to_multiple_of": 64, |
| 150 | "pad_token_id": 32000, |
| 151 | "text_config": { |
| 152 | "model_type": "llama", |
| 153 | "pad_token_id": 32000, |
| 154 | "torch_dtype": "bfloat16", |
| 155 | "vocab_size": 32064 |
| 156 | }, |
| 157 | "timm_model_ids": [ |
| 158 | "vit_large_patch14_reg4_dinov2.lvd142m", |
| 159 | "vit_so400m_patch14_siglip_224" |
| 160 | ], |
| 161 | "timm_override_act_layers": [ |
| 162 | null, |
| 163 | null |
| 164 | ], |
| 165 | "torch_dtype": "bfloat16", |
| 166 | "transformers_version": "4.40.1", |
| 167 | "use_fused_vision_backbone": true, |
| 168 | "vision_backbone_id": "dinosiglip-vit-so-224px" |
| 169 | } |
| 170 | |