config.json

{
  "architectures": [
    "GotOcr2ForConditionalGeneration"
  ],
  "ignore_index": -100,
  "image_seq_length": 576,
  "image_token_index": 151859,
  "model_type": "got_ocr2",
  "quantization_config": {
    "config_groups": {
      "group_0": {
        "input_activations": {
          "actorder": null,
          "block_structure": null,
          "dynamic": true,
          "group_size": null,
          "num_bits": 8,
          "observer": null,
          "observer_kwargs": {},
          "strategy": "token",
          "symmetric": true,
          "type": "float"
        },
        "output_activations": null,
        "targets": [
          "Linear"
        ],
        "weights": {
          "actorder": null,
          "block_structure": null,
          "dynamic": false,
          "group_size": null,
          "num_bits": 8,
          "observer": "minmax",
          "observer_kwargs": {},
          "strategy": "channel",
          "symmetric": true,
          "type": "float"
        }
      }
    },
    "format": "float-quantized",
    "global_compression_ratio": null,
    "ignore": [
      "model.vision_tower.layers.0.attn.qkv",
      "model.vision_tower.layers.0.attn.proj",
      "model.vision_tower.layers.0.mlp.lin1",
      "model.vision_tower.layers.0.mlp.lin2",
      "model.vision_tower.layers.1.attn.qkv",
      "model.vision_tower.layers.1.attn.proj",
      "model.vision_tower.layers.1.mlp.lin1",
      "model.vision_tower.layers.1.mlp.lin2",
      "model.vision_tower.layers.2.attn.qkv",
      "model.vision_tower.layers.2.attn.proj",
      "model.vision_tower.layers.2.mlp.lin1",
      "model.vision_tower.layers.2.mlp.lin2",
      "model.vision_tower.layers.3.attn.qkv",
      "model.vision_tower.layers.3.attn.proj",
      "model.vision_tower.layers.3.mlp.lin1",
      "model.vision_tower.layers.3.mlp.lin2",
      "model.vision_tower.layers.4.attn.qkv",
      "model.vision_tower.layers.4.attn.proj",
      "model.vision_tower.layers.4.mlp.lin1",
      "model.vision_tower.layers.4.mlp.lin2",
      "model.vision_tower.layers.5.attn.qkv",
      "model.vision_tower.layers.5.attn.proj",
      "model.vision_tower.layers.5.mlp.lin1",
      "model.vision_tower.layers.5.mlp.lin2",
      "model.vision_tower.layers.6.attn.qkv",
      "model.vision_tower.layers.6.attn.proj",
      "model.vision_tower.layers.6.mlp.lin1",
      "model.vision_tower.layers.6.mlp.lin2",
      "model.vision_tower.layers.7.attn.qkv",
      "model.vision_tower.layers.7.attn.proj",
      "model.vision_tower.layers.7.mlp.lin1",
      "model.vision_tower.layers.7.mlp.lin2",
      "model.vision_tower.layers.8.attn.qkv",
      "model.vision_tower.layers.8.attn.proj",
      "model.vision_tower.layers.8.mlp.lin1",
      "model.vision_tower.layers.8.mlp.lin2",
      "model.vision_tower.layers.9.attn.qkv",
      "model.vision_tower.layers.9.attn.proj",
      "model.vision_tower.layers.9.mlp.lin1",
      "model.vision_tower.layers.9.mlp.lin2",
      "model.vision_tower.layers.10.attn.qkv",
      "model.vision_tower.layers.10.attn.proj",
      "model.vision_tower.layers.10.mlp.lin1",
      "model.vision_tower.layers.10.mlp.lin2",
      "model.vision_tower.layers.11.attn.qkv",
      "model.vision_tower.layers.11.attn.proj",
      "model.vision_tower.layers.11.mlp.lin1",
      "model.vision_tower.layers.11.mlp.lin2",
      "lm_head"
    ],
    "kv_cache_scheme": null,
    "quant_method": "compressed-tensors",
    "quantization_status": "compressed"
  },
  "text_config": {
    "attention_dropout": 0.0,
    "hidden_act": "silu",
    "hidden_size": 1024,
    "initializer_range": 0.02,
    "intermediate_size": 2816,
    "max_position_embeddings": 32768,
    "max_window_layers": 21,
    "model_type": "qwen2",
    "num_attention_heads": 16,
    "num_hidden_layers": 24,
    "num_key_value_heads": 16,
    "rms_norm_eps": 1e-06,
    "rope_scaling": null,
    "rope_theta": 1000000.0,
    "sliding_window": 4096,
    "tie_word_embeddings": true,
    "torch_dtype": "bfloat16",
    "use_cache": true,
    "use_sliding_window": false,
    "vocab_size": 151860
  },
  "torch_dtype": "bfloat16",
  "transformers_version": "4.52.4",
  "use_cache": true,
  "vision_config": {
    "attention_dropout": 0.0,
    "global_attn_indexes": [
      2,
      5,
      8,
      11
    ],
    "hidden_act": "gelu",
    "hidden_size": 768,
    "image_size": 1024,
    "initializer_range": 1e-10,
    "layer_norm_eps": 1e-06,
    "mlp_dim": 3072,
    "model_type": "",
    "num_attention_heads": 12,
    "num_channels": 3,
    "num_hidden_layers": 12,
    "output_channels": 256,
    "patch_size": 16,
    "qkv_bias": true,
    "torch_dtype": "bfloat16",
    "use_abs_pos": true,
    "use_rel_pos": true,
    "window_size": 14
  }
}