config.json

{
  "additional_vocab_size": 40,
  "architectures": [
    "ColModernVBert"
  ],
  "auto_map": {
    "AutoConfig": "configuration_modernvbert.ModernVBertConfig",
    "AutoModel": "modeling_modernvbert.ModernVBertModel",
    "AutoModelForMaskedLM": "modeling_modernvbert.ModernVBertForMaskedLM"
  },
  "embedding_dim": 128,
  "initializer_range": 0.02,
  "vlm_config": {
    "image_token_id": 50407,
    "initializer_range": 0.02,
    "model_type": "modernvbert",
    "pixel_shuffle_factor": 4,
    "text_config": {
      "_name_or_path": "ettin-encoder-150m",
      "architectures": [
        "ModernBertForMaskedLM"
      ],
      "attention_bias": false,
      "attention_dropout": 0.0,
      "causal_mask": false,
      "classifier_activation": "gelu",
      "classifier_bias": false,
      "classifier_dropout": 0.0,
      "classifier_pooling": "mean",
      "cls_token_id": 50281,
      "decoder_bias": true,
      "deterministic_flash_attn": false,
      "dtype": "float32",
      "embedding_dropout": 0.0,
      "global_attn_every_n_layers": 3,
      "global_rope_theta": 160000.0,
      "gradient_checkpointing": false,
      "hidden_activation": "gelu",
      "hidden_size": 768,
      "initializer_cutoff_factor": 2.0,
      "initializer_range": 0.02,
      "intermediate_size": 1152,
      "is_causal": false,
      "layer_norm_eps": 1e-05,
      "layer_types": [
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention"
      ],
      "local_attention": 128,
      "local_rope_theta": 160000.0,
      "max_position_embeddings": 7999,
      "mlp_bias": false,
      "mlp_dropout": 0.0,
      "model_type": "modernbert",
      "norm_bias": false,
      "norm_eps": 1e-05,
      "num_attention_heads": 12,
      "num_hidden_layers": 22,
      "position_embedding_type": "sans_pos",
      "repad_logits_with_grad": false,
      "rope_parameters": {
        "full_attention": {
          "rope_theta": 160000.0,
          "rope_type": "default"
        },
        "sliding_attention": {
          "rope_theta": 160000.0,
          "rope_type": "default"
        }
      },
      "sparse_pred_ignore_index": -100,
      "sparse_prediction": false,
      "vocab_size": 50408
    },
    "transformers_version": "5.0.0.dev0",
    "vision_config": {
      "attention_dropout": 0.0,
      "hidden_act": "gelu_pytorch_tanh",
      "hidden_size": 768,
      "image_size": 512,
      "intermediate_size": 3072,
      "layer_norm_eps": 1e-06,
      "model_type": "siglip_vision_model",
      "num_attention_heads": 12,
      "num_channels": 3,
      "num_hidden_layers": 12,
      "patch_size": 16
    },
    "tie_word_embeddings": false
  }
}