config.json

{
    "_name_or_path": "openbmb/CPM-2B",
    "architectures": [
        "MiniCPMForCausalLM"
    ],
    "auto_map": {
        "AutoConfig": "configuration_minicpm.MiniCPMConfig",
        "AutoModel": "modeling_minicpm.MiniCPMModel",
        "AutoModelForCausalLM": "modeling_minicpm.MiniCPMForCausalLM",
        "AutoModelForSeq2SeqLM": "modeling_minicpm.MiniCPMForCausalLM",
        "AutoModelForSequenceClassification": "modeling_minicpm.MiniCPMForSequenceClassification"
    },
    "bos_token_id": 1,
    "eos_token_id": 2,
    "hidden_act": "silu",
    "hidden_size": 2304,
    "initializer_range": 0.1,
    "intermediate_size": 5760,
    "max_position_embeddings": 2048,
    "num_attention_heads": 36,
    "num_hidden_layers": 40,
    "num_key_value_heads": 36,
    "rms_norm_eps": 1e-05,
    "rope_scaling": null,
    "torch_dtype": "float16",
    "transformers_version": "4.36.0",
    "use_cache": true,
    "vocab_size": 122753,
    "scale_emb": 12,
    "dim_model_base": 256,
    "scale_depth": 1.4,
    "quantization_config": {
        "bits": 4,
        "group_size": 32,
        "damp_percent": 0.01,
        "desc_act": true,
        "sym": false,
        "true_sequential": true,
        "model_name_or_path": null,
        "model_file_base_name": null,
        "quant_method": "gptq",
        "modules_in_block_to_quantize": [
            [
                "self_attn.k_proj",
                "self_attn.v_proj",
                "self_attn.q_proj"
            ],
            [
                "self_attn.o_proj"
            ],
            [
                "mlp.up_proj",
                "mlp.gate_proj"
            ],
            [
                "mlp.down_proj"
            ]
        ]
    }
}