大模型实战营第二次课作业

发表于 2024-01-06 更新于 2024-03-19

基础作业

使用 InternLM-Chat-7B 模型生成 300 字的小故事

Machines help machines (笑)

使用 `huggingface_hub` python 包，下载 `InternLM-20B` 的 `config.json` 文件到本地

代码

1 2	from huggingface_hub import hf_hub_download hf_hub_download(repo_id="internlm/internlm-20b", filename="config.json")

下载过程

文件内容

{
  "architectures": [
    "InternLMForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "configuration_internlm.InternLMConfig",
    "AutoModel": "modeling_internlm.InternLMForCausalLM",
    "AutoModelForCausalLM": "modeling_internlm.InternLMForCausalLM"
  },
  "bias": false,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 13824,
  "max_position_embeddings": 4096,
  "model_type": "internlm",
  "num_attention_heads": 40,
  "num_hidden_layers": 60,
  "num_key_value_heads": 40,
  "pad_token_id": 2,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.33.1",
  "use_cache": true,
  "vocab_size": 103168,
  "rotary": {
    "base": 10000,
    "type": "dynamic"
  }
}