{
  "adapter_layer_norm_eps": 1e-06,
  "architectures": [
    "AyaVisionForConditionalGeneration"
  ],
  "bos_token_id": 0,
  "downsample_factor": 2,
  "dtype": "float32",
  "eos_token_id": 0,
  "image_token_index": 2,
  "model_type": "aya_vision",
  "pad_token_id": 0,
  "text_config": {
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 0,
    "eos_token_id": 0,
    "head_dim": 32,
    "hidden_act": "silu",
    "hidden_size": 128,
    "initializer_range": 0.02,
    "intermediate_size": 37,
    "layer_norm_eps": 1e-05,
    "layer_types": [
      "sliding_attention",
      "sliding_attention"
    ],
    "logit_scale": 0.0625,
    "max_position_embeddings": 512,
    "model_type": "cohere2",
    "num_attention_heads": 4,
    "num_hidden_layers": 2,
    "num_key_value_heads": 4,
    "output_channels": 64,
    "pad_token_id": 0,
    "rope_parameters": {
      "rope_theta": 10000.0,
      "rope_type": "default"
    },
    "sliding_window": 4096,
    "tie_word_embeddings": true,
    "use_cache": true,
    "vocab_size": 1024
  },
  "tie_word_embeddings": true,
  "transformers_version": "5.15.0.dev0",
  "vision_config": {
    "attention_dropout": 0.0,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 32,
    "image_size": 64,
    "intermediate_size": 128,
    "layer_norm_eps": 1e-06,
    "model_type": "siglip_vision_model",
    "num_attention_heads": 4,
    "num_channels": 3,
    "num_hidden_layers": 2,
    "patch_size": 8,
    "vision_use_head": false
  },
  "vision_feature_layer": -1,
  "vision_feature_select_strategy": "full"
}