README.md

---
license: gpl-3.0
language:
- en
datasets:
- Mxode/Magpie-Pro-10K-GPT4o-mini
pipeline_tag: text2text-generation
---
# NanoLM-25M-Instruct-v1.1


English | [简体中文](README_zh-CN.md)


## Introduction

In order to explore the potential of small models, I have attempted to build a series of them, which are available in the [NanoLM Collections](https://huggingface.co/collections/Mxode/nanolm-66d6d75b4a69536bca2705b2).

This is NanoLM-25M-Instruct-v1.1. The model currently supports **English only**.



## Model Details

| 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|



## How to use

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_path = 'Mxode/NanoLM-25M-Instruct-v1.1'

model = AutoModelForCausalLM.from_pretrained(model_path).to('cuda:0', torch.bfloat16)
tokenizer = AutoTokenizer.from_pretrained(model_path)


def get_response(prompt: str, **kwargs):
    generation_args = dict(
        max_new_tokens = kwargs.pop("max_new_tokens", 512),
        do_sample = kwargs.pop("do_sample", True),
        temperature = kwargs.pop("temperature", 0.7),
        top_p = kwargs.pop("top_p", 0.8),
        top_k = kwargs.pop("top_k", 40),
        **kwargs
    )

    messages = [
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": prompt}
    ]
    text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True
    )
    model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

    generated_ids = model.generate(model_inputs.input_ids, **generation_args)
    generated_ids = [
        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
    ]

    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
    return response


prompt1 = "What can you do for me?"
print(get_response(prompt1, do_sample=False))

"""
I'm so glad you asked! I'm a large language model, so I don't have personal experiences or emotions, but I can provide information and assist with tasks to help with your tasks.

Here are some ways I can assist you:

1. **Answer questions**: I can provide information on a wide range of topics, from science and history to entertainment and culture.
2. **Generate text**: I can create text based on a prompt or topic, and can even help with writing tasks such as proofreading and editing.
3. **Translate text**: I can translate text from one language to another, including popular languages such as Spanish, French, German, Chinese, and many more.
4. **Summarize content**: I can summarize long pieces of text, such as articles or documents, into shorter, more digestible versions.
5. **Offer suggestions**: I can provide suggestions for things like gift ideas, travel destinations, books, or movies.
6. **Chat and converse**: I can engage in natural-sounding conversations, using context and understanding to respond to questions and statements.
7. **Play games**: I can play simple text-based games, such as 20 Questions, Hangman, or Word Jumble.
8. **Provide definitions**: I can define words and phrases, explaining their meanings and usage.
9. **Offer suggestions**: I can provide suggestions for things like gift ideas, travel destinations, or books to read.
10. **Entertain**: I can engage in fun conversations, tell jokes, and even create simple games or puzzles.

Which of these methods would you like to do?
"""
```