config.json

{
  "_name_or_path": "THUDM/glm-4v-9b",
  "model_type": "chatglm",
  "architectures": [
    "ChatGLMModel"
  ],
  "auto_map": {
    "AutoConfig": "configuration_chatglm.ChatGLMConfig",
    "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
    "AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
    "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
    "AutoModelForSequenceClassification": "modeling_chatglm.ChatGLMForSequenceClassification"
  },
  "vision_config": {
    "dropout_prob": 0.0,
    "hidden_act": "gelu",
    "in_channels": 3,
    "num_hidden_layers": 63,
    "hidden_size": 1792,
    "patch_size": 14,
    "num_heads": 16,
    "intermediate_size": 15360,
    "layer_norm_eps": 1e-06,
    "num_positions": 6401,
    "image_size": 1120,
    "scaling_factor": 8
  },
  "add_bias_linear": false,
  "add_qkv_bias": true,
  "apply_query_key_layer_scaling": true,
  "apply_residual_connection_post_layernorm": false,
  "attention_dropout": 0.0,
  "attention_softmax_in_fp32": true,
  "attn_implementation": "sdpa",
  "bias_dropout_fusion": true,
  "ffn_hidden_size": 13696,
  "fp32_residual_connection": false,
  "hidden_dropout": 0.0,
  "hidden_size": 4096,
  "kv_channels": 128,
  "layernorm_epsilon": 0.00000015625,
  "multi_query_attention": true,
  "multi_query_group_num": 2,
  "num_attention_heads": 32,
  "num_layers": 40,
  "original_rope": true,
  "padded_vocab_size": 151552,
  "post_layer_norm": true,
  "rmsnorm": true,
  "seq_length": 8192,
  "use_cache": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.0",
  "tie_word_embeddings": false,
  "eos_token_id": [151329, 151336, 151338],
  "pad_token_id": 151329,
  "boi_token_id": 151339,
  "eoi_token_id": 151340,
  "quantization_config": {
    "bits": 4,
    "group_size": 128,
    "damp_percent": 0.01,
    "desc_act": false,
    "static_groups": false,
    "sym": true,
    "true_sequential": true,
    "model_name_or_path": null,
    "model_file_base_name": null,
    "quant_method": "gptq",
    "checkpoint_format": "gptq",
    "block_name_to_quantize": "",
    "modules_in_block_to_quantize": [
        ["self_attention.query_key_value", "self_attention.dense", "mlp.dense_h_to_4h", "mlp.dense_4h_to_h"],
        ["attention.query_key_value", "attention.dense", "mlp.fc1", "mlp.fc2"],
        ["linear_proj", "dense_h_to_4h", "gate_proj", "dense_4h_to_h"]
    ],
    "disable_exllama": true
  }
}