config.json

{
  "architectures": [
    "LlavaForCausalLM"
  ],
  "model_type": "llava",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.36.0",
  "_name_or_path": "DeepXR/Helion-V2.0-Thinking",
  "vision_config": {
    "hidden_size": 1152,
    "image_size": 384,
    "intermediate_size": 4304,
    "model_type": "siglip_vision_model",
    "num_attention_heads": 16,
    "num_hidden_layers": 27,
    "patch_size": 14,
    "projection_dim": 768,
    "vocab_size": 32000
  },
  "text_config": {
    "architectures": [
      "LlamaForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 1,
    "eos_token_id": 2,
    "hidden_act": "silu",
    "hidden_size": 4096,
    "initializer_range": 0.02,
    "intermediate_size": 14336,
    "max_position_embeddings": 200000,
    "model_type": "llama",
    "num_attention_heads": 32,
    "num_hidden_layers": 48,
    "num_key_value_heads": 8,
    "pretraining_tp": 1,
    "rms_norm_eps": 1e-05,
    "rope_scaling": {
      "type": "linear",
      "factor": 8.0
    },
    "rope_theta": 500000.0,
    "tie_word_embeddings": false,
    "torch_dtype": "bfloat16",
    "use_cache": true,
    "vocab_size": 32000,
    "pad_token_id": 0
  },
  "image_token_index": 32000,
  "vision_feature_select_strategy": "default",
  "vision_feature_layer": -2,
  "use_flash_attention_2": true,
  "ignore_index": -100,
  "image_seq_length": 729,
  "vocab_size": 32001,
  "projector_hidden_act": "gelu",
  "vision_aspect_ratio": "anyres",
  "image_grid_pinpoints": [
    [336, 672],
    [672, 336],
    [672, 672],
    [1008, 336],
    [336, 1008]
  ]
}