Pinokio V1.0

🚀 Pinokio_v1.0

Pinokio_v1.0 is a model in the transformers library, which can be used for text - generation tasks.

🚀 Quick Start

To start using the Pinokio_v1.0 model, you need to install the necessary libraries first. Then, you can follow the steps below to perform text - generation tasks.

💻 Usage Examples

Basic Usage

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Eurdem/Pinokio_v1.0"

#tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    model_kwargs={"torch_dtype": torch.bfloat16, "load_in_8bit": True},
)

messages = [{"role": "user", "content": """Her gün 30 km koşarsam, 270 km yolu kaç günde koşabilirim?"""}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, do_sample=True, temperature=0.7, top_k=500, top_p=0.7,max_new_tokens=1024)
print(outputs[0]["generated_text"])

Advanced Usage

The basic usage above provides a simple way to use the model. For more advanced scenarios, you can adjust parameters such as temperature, top_k, top_p according to your needs to get different text - generation results. For example, you can try different values of temperature to control the randomness of the generated text. A higher temperature value will make the generated text more random, while a lower value will make it more deterministic.

# You can adjust the parameters according to your needs
outputs = pipeline(prompt, do_sample=True, temperature=0.9, top_k=600, top_p=0.8,max_new_tokens=1500)
print(outputs[0]["generated_text"])

📄 License

This project is licensed under the Apache - 2.0 license.