OuteAI_-_Lite-Oute-1-300M-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Lite-Oute-1-300M - GGUF

• Model creator: https://huggingface.co/OuteAI/
• Original model: https://huggingface.co/OuteAI/Lite-Oute-1-300M/

Original model description:

license: apache-2.0

Lite-Oute-1-300M

Lite-Oute-1-300M (Base) is a Lite series model based on the Mistral architecture, comprising approximately 300 million parameters. <br> This model is specifically designed as a starting point for fine-tuning on various tasks. With its 300 million parameters, it offers a balance between compact size and capability, making it suitable for a wide range of fine-tuning applications.<br> The model was trained on 30 billion tokens with a context length of 4096, providing a solid foundation for task-specific adaptations.

Available versions:

<a href="https://huggingface.co/OuteAI/Lite-Oute-1-300M-Instruct">Lite-Oute-1-300M-Instruct</a> <br> <a href="https://huggingface.co/OuteAI/Lite-Oute-1-300M-Instruct-GGUF">Lite-Oute-1-300M-Instruct-GGUF</a> <br> <a href="https://huggingface.co/OuteAI/Lite-Oute-1-300M-Instruct">Lite-Oute-1-300M</a> <br> <a href="https://huggingface.co/OuteAI/Lite-Oute-1-300M-Instruct-GGUF">Lite-Oute-1-300M-GGUF</a> <br>

Benchmarks:

<table style="text-align: left;"> <tr> <th>Benchmark</th> <th>5-shot</th> <th>0-shot</th> </tr> <tr> <td>ARC Challenge</td> <td>26.62</td> <td>26.28</td> </tr> <tr> <td>ARC Easy</td> <td>51.39</td> <td>48.11</td> </tr> <tr> <td>CommonsenseQA</td> <td>19.49</td> <td>20.64</td> </tr> <tr> <td>HellaSWAG</td> <td>34.86</td> <td>34.85</td> </tr> <tr> <td>MMLU</td> <td>27.23</td> <td>24.87</td> </tr> <tr> <td>OpenBookQA</td> <td>30.20</td> <td>30.80</td> </tr> <tr> <td>PIQA</td> <td>65.07</td> <td>65.02</td> </tr> <tr> <td>Winogrande</td> <td>51.14</td> <td>53.35</td> </tr> </table>

Usage with HuggingFace transformers

The model can be used with HuggingFace's transformers library:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

model = AutoModelForCausalLM.from_pretrained("OuteAI/Lite-Oute-1-300M").to(device)
tokenizer = AutoTokenizer.from_pretrained("OuteAI/Lite-Oute-1-300M")

def generate_response(message: str, temperature: float = 0.4, repetition_penalty: float = 1.12) -> str:
    # Convert message to PyTorch tensors
    input_ids = tokenizer.encode(
        message, return_tensors="pt"
    ).to(device)
    # Generate the response
    output = model.generate(
        input_ids,
        max_length=256,
        temperature=temperature,
        repetition_penalty=repetition_penalty,
        do_sample=True
    ) 
    # Decode the generated output
    generated_text = tokenizer.decode(output[0], skip_special_tokens=True)
    return generated_text
message = "Scientists have made a breakthrough in renewable energy by developing a new type of"
response = generate_response(message)
print(response)

Risk Disclaimer

By using this model, you acknowledge that you understand and assume the risks associated with its use. You are solely responsible for ensuring compliance with all applicable laws and regulations. We disclaim any liability for problems arising from the use of this open-source model, including but not limited to direct, indirect, incidental, consequential, or punitive damages. We make no warranties, express or implied, regarding the model's performance, accuracy, or fitness for a particular purpose. Your use of this model is at your own risk, and you agree to hold harmless and indemnify us, our affiliates, and our contributors from any claims, damages, or expenses arising from your use of the model.