config.json

{
  "architectures": [
    "SpatialLMQwenForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 896,
  "initializer_range": 0.02,
  "intermediate_size": 4864,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "spatiallm_qwen",
  "num_attention_heads": 14,
  "num_hidden_layers": 24,
  "num_key_value_heads": 2,
  "point_backbone": "sonata",
  "point_config": {
    "enc_channels": [
      48,
      96,
      192,
      384,
      512
    ],
    "enc_depths": [
      3,
      3,
      3,
      12,
      3
    ],
    "enc_mode": true,
    "enc_num_head": [
      3,
      6,
      12,
      24,
      32
    ],
    "enc_patch_size": [
      1024,
      1024,
      1024,
      1024,
      1024
    ],
    "in_channels": 6,
    "mask_token": true,
    "mlp_ratio": 4,
    "order": [
      "z",
      "z-trans",
      "hilbert",
      "hilbert-trans"
    ],
    "stride": [
      2,
      2,
      2,
      2
    ],
    "num_bins": 1280
  },
  "projector": "mlp",
  "point_end_token_id": 151653,
  "point_start_token_id": 151652,
  "point_token_id": 151654,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "torch_dtype": "float32",
  "transformers_version": "4.46.1",
  "use_cache": false,
  "use_sliding_window": false,
  "vocab_size": 151936
}