{
  "architectures": [
    "VisualRelatednessModel"
  ],
  "model_type": "bert-cnn-caption-reranker",
  "bert_config": {
    "return_dict": true,
    "output_hidden_states": false,
    "dtype": "float32",
    "chunk_size_feed_forward": 0,
    "is_encoder_decoder": false,
    "architectures": [
      "BertForMaskedLM"
    ],
    "id2label": {
      "0": "LABEL_0",
      "1": "LABEL_1"
    },
    "label2id": {
      "LABEL_0": 0,
      "LABEL_1": 1
    },
    "problem_type": null,
    "_name_or_path": "models/bert-base-uncased",
    "transformers_version": "5.3.0",
    "gradient_checkpointing": false,
    "model_type": "bert",
    "position_embedding_type": "absolute",
    "pad_token_id": 0,
    "is_decoder": false,
    "add_cross_attention": false,
    "bos_token_id": null,
    "eos_token_id": null,
    "tie_word_embeddings": true,
    "vocab_size": 30522,
    "hidden_size": 768,
    "num_hidden_layers": 12,
    "num_attention_heads": 12,
    "hidden_act": "gelu",
    "intermediate_size": 3072,
    "hidden_dropout_prob": 0.1,
    "attention_probs_dropout_prob": 0.1,
    "max_position_embeddings": 512,
    "type_vocab_size": 2,
    "initializer_range": 0.02,
    "layer_norm_eps": 1e-12,
    "use_cache": true,
    "classifier_dropout": null,
    "output_attentions": false
  },
  "model_options": {
    "architecture": "bert-cnn",
    "seq_len": 64,
    "cnn_filters": 32,
    "cnn_kernel_size": 3,
    "num_bert_layers": 12,
    "tune_embeddings": false
  },
  "rank_prior": {
    "bias": [
      2.5842955112457275,
      -0.21887986361980438,
      -2.617722511291504,
      -3.491527795791626,
      -4.005457878112793,
      -3.7829275131225586,
      -5.263741970062256,
      -5.885319709777832,
      -6.846593379974365
    ],
    "weight": 3.2007904052734375
  },
  "score": "bias[beam position] + weight * BERT-CNN logit; beam positions 1..9 in beam-search order"
}