config.json

{
"architectures": [
"VisionEncoderDecoderModel"
],
"decoder": {
"bos_token_id": 50256,
"decoder_start_token_id": 50256,
"eos_token_id": 50256,
"hidden_size": 768,
"intermediate_size": 3072,
"layer_norm_eps": 1e-05,
"max_length": 200,
"model_type": "gpt2",
"num_attention_heads": 12,
"num_hidden_layers": 12,
"pad_token_id": 50256,
"transformers_version": "4.35.0",
"vocab_size": 50257
},
"encoder": {
"hidden_act": "gelu",
"hidden_size": 768,
"image_size": 224,
"intermediate_size": 3072,
"layer_norm_eps": 1e-12,
"model_type": "vit",
"num_attention_heads": 12,
"num_hidden_layers": 12,
"patch_size": 16,
"qkv_bias": true,
"transformers_version": "4.35.0"
},
"is_encoder_decoder": true,
"tie_word_embeddings": false,
"torch_dtype": "float32"
}