config.json

{
    "_name_or_path": "/workspace/process/opennlplab_transnormerllm-7b/source",
    "add_bos_token": false,
    "architectures": [
        "TransnormerForCausalLM"
    ],
    "auto_map": {
        "AutoConfig": "configuration_transnormer.TransnormerConfig",
        "AutoModelForCausalLM": "modeling_transnormer.TransnormerForCausalLM"
    },
    "bias": false,
    "bos_token_id": 1,
    "decoder_attention_heads": 32,
    "decoder_embed_dim": 4096,
    "decoder_layers": 30,
    "eos_token_id": 2,
    "glu_dim": 11008,
    "hidden_dim": 4096,
    "init_std": 0.02,
    "linear_act_fun": "swish",
    "linear_use_lrpe": 0,
    "linear_use_lrpe_list": [
        1,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        0
    ],
    "model_max_length": 8192,
    "model_type": "transnormer",
    "no_scale_embedding": true,
    "norm_type": "simplermsnorm",
    "pad_token_id": 0,
    "pretraining_tp": 1,
    "quantization_config": {
        "batch_size": 1,
        "bits": 4,
        "block_name_to_quantize": "model.layers",
        "cache_block_outputs": true,
        "damp_percent": 0.1,
        "desc_act": true,
        "exllama_config": {
            "version": 1
        },
        "group_size": 128,
        "max_input_length": null,
        "model_seqlen": 8192,
        "module_name_preceding_first_block": [
            "model.embed_tokens"
        ],
        "pad_token_id": null,
        "quant_method": "gptq",
        "sym": true,
        "tokenizer": null,
        "true_sequential": true,
        "use_cuda_fp16": false,
        "use_exllama": true
    },
    "torch_dtype": "bfloat16",
    "transformers_version": "4.35.0",
    "use_cache": true,
    "vocab_size": 64000
}