llama3的改進

llama2
{
"_name_or_path": "TheBloke/Llama-2-7B-fp16",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 11008,
"max_position_embeddings": 2048,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"pad_token_id": 0,
"rms_norm_eps": 1e-05,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.30.2",
"use_cache": true,
"vocab_size": 32000
}

llama3:
{
"_name_or_path": "../../llama-3-8b",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 128000,
"eos_token_id": 128001,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 14336,
"max_position_embeddings": 8192,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 500000.0,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.39.3",
"use_cache": true,
"vocab_size": 128256
}

區別:
"bos_token_id": 128000, 跟2不一樣
"eos_token_id": 128001, 跟2不一樣
"intermediate_size": 14336,比2大
"max_position_embeddings": 8192,比2大 2只有2k說明句子長度變成8k了.
"torch_dtype": "bfloat16", 2代用的32位來存儲的. 說明16位模型是更好的效率的方案.
"transformers_version": "4.38.2", 需要的transformers的版本也更高了.
"vocab_size": 128256 #添加了大量的vocab_size 之前只有3w2
"num_key_value_heads": 8, key value 的維度比q要小四倍.

整體上都是維度上的微調和字典增加.

ps

可以看到翻了這個vocab, 一箇中文都沒有
編碼的例子:

具體中文能力還是看測試吧.