65 lines
2.0 KiB
Markdown
65 lines
2.0 KiB
Markdown
# NanoLM-70M-Instruct-v1
|
||
|
||
[English](README.md) | 简体中文
|
||
|
||
|
||
## Introduction
|
||
|
||
为了探究小模型的潜能，我尝试构建一系列小模型，并存放于 [NanoLM Collections](https://huggingface.co/collections/Mxode/nanolm-66d6d75b4a69536bca2705b2)。
|
||
|
||
这是 NanoLM-70M-Instruct-v1。该模型目前仅支持**英文**。
|
||
|
||
|
||
## 模型详情
|
||
|
||
| Nano LMs | Non-emb Params | Arch | Layers | Dim | Heads | Seq Len |
|
||
| :----------: | :------------------: | :---: | :----: | :-------: | :---: | :---: |
|
||
| 25M | 15M | MistralForCausalLM | 12 | 312 | 12 |2K|
|
||
| **70M** | **42M** | **LlamaForCausalLM** | **12** | **576** | **9** | **2K** |
|
||
| 0.3B | 180M | Qwen2ForCausalLM | 12 | 896 | 14 |4K|
|
||
| 1B | 840M | Qwen2ForCausalLM | 18 | 1536 | 12 |4K|
|
||
|
||
NanoLM-70M-Instruct-v1 的分词器和模型架构与 [SmolLM-135M](https://huggingface.co/HuggingFaceTB/SmolLM-135M) 相同，但层数从30减少到12。
|
||
|
||
本质上是纯粹的 LLaMA 架构，即 LlamaForCausalLM。
|
||
|
||
因此，NanoLM-70M-Instruct-v1 的参数量只有 70 M。
|
||
|
||
尽管如此，NanoLM-70M-Instruct-v1 仍展示了指令跟随能力。
|
||
|
||
|
||
## 如何使用
|
||
|
||
```python
|
||
import torch
|
||
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
||
model_path = 'Mxode/NanoLM-70M-Instruct-v1'
|
||
|
||
model = AutoModelForCausalLM.from_pretrained(model_path).to('cuda:0', torch.bfloat16)
|
||
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
||
|
||
|
||
text = "Why is it important for entrepreneurs to prioritize financial management?"
|
||
prompt = tokenizer.apply_chat_template(
|
||
[
|
||
{'role': 'system', 'content': 'You are a helpful assistant.'},
|
||
{'role': 'user', 'content': text}
|
||
],
|
||
add_generation_prompt=True,
|
||
tokenize=True,
|
||
return_tensors='pt'
|
||
).to('cuda:0')
|
||
|
||
|
||
outputs = model.generate(
|
||
prompt,
|
||
max_new_tokens=1024,
|
||
do_sample=True,
|
||
temperature=0.7,
|
||
repetition_penalty=1.1,
|
||
eos_token_id=tokenizer.eos_token_id,
|
||
)
|
||
response = tokenizer.decode(outputs[0])
|
||
print(response)
|
||
``` |