Model(
  (embed_tokens): Embedding(32000, 4096, padding_idx=0)
  (layers): ModuleList(
    (0): LlamaDecoderLayer(
      (self_attn): LlamaAttention(
        (q_proj): Linear(in_features=4096, out_features=4096, bias=False)
        (k_proj): Linear(in_features=4096, out_features=4096, bias=False)
        (v_proj): Linear(in_features=4096, out_features=4096, bias=False)
        (o_proj): Linear(in_features=4096, out_features=4096, bias=False)
        (rotary_emb): LlamaRotaryEmbedding()
      )
      (mlp): LlamaMLP(
        (gate_proj): Linear(in_features=4096, out_features=11008, bias=False)
        (up_proj): Linear(in_features=4096, out_features=11008, bias=False)
        (down_proj): Linear(in_features=11008, out_features=4096, bias=False)
        (act_fn): SiLU()
      )
      (post_attention_layernorm): LlamaRMSNorm()
    )
  )
  (fc): Linear(in_features=8192, out_features=4096, bias=True)
  (act): SiLU()
)

vicuna-13B-config.json

json

{
  "architectures": [
    "LlamaForCausalLM"
  ],
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 13824,
  "max_position_embeddings": 2048,
  "model_type": "llama",
  "num_attention_heads": 40,
  "num_hidden_layers": 1,
  "pad_token_id": 0,
  "rms_norm_eps": 1e-06,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.28.1",
  "use_cache": true,
  "vocab_size": 32000
}

模型训练

数据生成

python -m eagle.ge_data.allocation --outdir ../data

训练

accelerate launch -m --mixed_precision=bf16 eagle.train.main --tmpdir eagle/data/sharegpt_0_67999_mufp16 --cpdir eagle/checkpoint --configpath eagle/train/vicuna_13B_config.json

模型上传

from huggingface_hub import HfApi

api = HfApi()

# 只上传修改后的 README.md 文件
api.upload_file(
    path_or_fileobj="checkpoints/eagle-vicuna-13B/README.md",  # 本地修改后的 README 路径
    path_in_repo="README.md",                                   # 仓库中的目标路径（根目录）
    repo_id="Gavin1104/eagle-vicuna-13b-v1.3",
    repo_type="model"
)