{
  "architectures": [
    "Blip2QFormerModel"
  ],
  "attention_dropout": 0.1,
  "attention_probs_dropout_prob": 0.1,
  "bos_token_id": 2,
  "cross_attention_frequency": 2,
  "dropout": 0.1,
  "dtype": "float32",
  "encoder_hidden_size": 32,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 32,
  "initializer_range": 0.02,
  "intermediate_size": 37,
  "layer_norm_eps": 1e-12,
  "max_position_embeddings": 512,
  "model_type": "blip_2_qformer",
  "num_attention_heads": 4,
  "num_hidden_layers": 2,
  "pad_token_id": 1,
  "projection_dim": 32,
  "transformers_version": "5.16.0.dev0",
  "use_qformer_text_input": false,
  "vocab_size": 50266
}