config.json

{
  "_name_or_path": "xvla",
  "model_type": "xvla",
  "architectures": ["XVLA"],
  "auto_map": {
    "AutoConfig": "configuration_xvla.XVLAConfig",
    "AutoModel": "modeling_xvla.XVLA"
  },

  "action_mode": "ee6d",
  "use_proprio": true,
  "num_actions": 30,

  "hidden_size": 1024,
  "depth": 24,
  "num_heads": 16,
  "mlp_ratio": 4.0,
  "num_domains": 30,
  "len_soft_prompts": 32,
  "dim_time": 32,
  "max_len_seq": 512,
  "use_hetero_proj": false,
  "soft_prompt_length": 32,

  "florence_config": {
    "model_type": "florence2",
    "bos_token_id": 0,
    "eos_token_id": 2,
    "ignore_index": -100,
    "pad_token_id": 1,
    "projection_dim": 1024,

    "text_config": {
      "vocab_size": 51289,
      "activation_dropout": 0.1,
      "activation_function": "gelu",
      "attention_dropout": 0.1,
      "d_model": 1024,
      "decoder_attention_heads": 16,
      "decoder_layers": 12,
      "encoder_attention_heads": 16,
      "encoder_layers": 12,
      "dropout": 0.1,
      "max_position_embeddings": 4096,
      "num_hidden_layers": 12,
      "num_beams": 3
    },

    "vision_config": {
      "model_type": "davit",
      "drop_path_rate": 0.1,
      "patch_size": [7, 3, 3, 3],
      "patch_stride": [4, 2, 2, 2],
      "patch_padding": [3, 1, 1, 1],
      "patch_prenorm": [false, true, true, true],
      "enable_checkpoint": false,
      "dim_embed": [256, 512, 1024, 2048],
      "num_heads": [8, 16, 32, 64],
      "num_groups": [8, 16, 32, 64],
      "depths": [1, 1, 9, 1],
      "window_size": 12,
      "projection_dim": 1024,
      "visual_temporal_embedding": {
        "type": "COSINE",
        "max_temporal_embeddings": 100
      },
      "image_pos_embed": {
        "type": "learned_abs_2d",
        "max_pos_embeddings": 50
      },
      "image_feature_source": ["spatial_avg_pool", "temporal_avg_pool"]
    },

    "vocab_size": 51289,
    "torch_dtype": "float32",
    "is_encoder_decoder": true
  }
}