config.json

{
  "architectures": [
    "Idefics2ForConditionalGeneration"
  ],
  "image_token_id": 32001,
  "model_type": "idefics2",
  "perceiver_config": {
    "attention_dropout": 0.0,
    "hidden_act": "silu",
    "hidden_size": 8,
    "initializer_range": 0.02,
    "model_type": "idefics2_perceiver",
    "num_key_value_heads": 4,
    "resampler_depth": 3,
    "resampler_head_dim": 96,
    "resampler_n_heads": 16,
    "resampler_n_latents": 64,
    "rms_norm_eps": 1e-06
  },
  "text_config": {
    "attention_dropout": 0.0,
    "head_dim": null,
    "hidden_act": "silu",
    "hidden_size": 8,
    "initializer_range": 0.02,
    "intermediate_size": 32,
    "max_position_embeddings": 131072,
    "model_type": "mistral",
    "num_attention_heads": 4,
    "num_hidden_layers": 2,
    "num_key_value_heads": 2,
    "rms_norm_eps": 1e-06,
    "rope_theta": 10000.0,
    "sliding_window": 4096,
    "use_cache": true,
    "vocab_size": 32006
  },
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.54.0.dev0",
  "use_cache": true,
  "vision_config": {
    "attention_dropout": 0.0,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 16,
    "image_size": 224,
    "initializer_range": 0.02,
    "intermediate_size": 32,
    "layer_norm_eps": 1e-06,
    "model_type": "idefics2_vision",
    "num_attention_heads": 4,
    "num_channels": 3,
    "num_hidden_layers": 2,
    "patch_size": 32
  }
}