{
  "architectures": [
    "VisionEncoderDecoderModel"
  ],
  "decoder": {
    "_name_or_path": "/tmp/tmp2lxjeaeg/decoder/GPT2LMHeadModel",
    "activation_function": "gelu_new",
    "add_cross_attention": true,
    "architectures": [
      "GPT2LMHeadModel"
    ],
    "attn_pdrop": 0.1,
    "bos_token_id": 50256,
    "chunk_size_feed_forward": 0,
    "dtype": "float32",
    "embd_pdrop": 0.1,
    "eos_token_id": 50256,
    "id2label": {
      "0": "LABEL_0",
      "1": "LABEL_1"
    },
    "initializer_range": 0.02,
    "is_decoder": true,
    "is_encoder_decoder": false,
    "label2id": {
      "LABEL_0": 0,
      "LABEL_1": 1
    },
    "layer_norm_epsilon": 1e-05,
    "model_type": "gpt2",
    "n_embd": 32,
    "n_head": 2,
    "n_inner": null,
    "n_layer": 2,
    "n_positions": 512,
    "output_attentions": false,
    "output_hidden_states": false,
    "pad_token_id": 0,
    "problem_type": null,
    "reorder_and_upcast_attn": false,
    "resid_pdrop": 0.1,
    "return_dict": true,
    "scale_attn_by_inverse_layer_idx": false,
    "scale_attn_weights": true,
    "summary_activation": null,
    "summary_first_dropout": 0.1,
    "summary_proj_to_labels": true,
    "summary_type": "cls_index",
    "summary_use_proj": true,
    "tie_word_embeddings": false,
    "use_cache": true,
    "vocab_size": 50257
  },
  "dtype": "float32",
  "encoder": {
    "_name_or_path": "/tmp/tmp2lxjeaeg/encoder/ViTModel",
    "architectures": [
      "ViTModel"
    ],
    "attention_probs_dropout_prob": 0.1,
    "chunk_size_feed_forward": 0,
    "dtype": "float32",
    "encoder_stride": 2,
    "hidden_act": "gelu",
    "hidden_dropout_prob": 0.1,
    "hidden_size": 32,
    "id2label": {
      "0": "LABEL_0",
      "1": "LABEL_1"
    },
    "image_size": 30,
    "initializer_range": 0.02,
    "intermediate_size": 37,
    "is_decoder": false,
    "is_encoder_decoder": false,
    "label2id": {
      "LABEL_0": 0,
      "LABEL_1": 1
    },
    "layer_norm_eps": 1e-12,
    "model_type": "vit",
    "num_attention_heads": 4,
    "num_channels": 3,
    "num_hidden_layers": 2,
    "output_attentions": false,
    "output_hidden_states": false,
    "patch_size": 2,
    "pooler_act": "tanh",
    "pooler_output_size": 32,
    "problem_type": null,
    "qkv_bias": true,
    "return_dict": true
  },
  "is_encoder_decoder": true,
  "model_type": "vision-encoder-decoder",
  "tie_word_embeddings": false,
  "transformers_version": "5.16.0.dev0"
}