---
license: mit
language:
- tr
pipeline_tag: text-generation

tags:
- Turkish
- turkish
- gpt2
- instruction-tuning
- alpaca
---


---
library_name: peft
base_model: ytu-ce-cosmos/turkish-gpt2-large
---

# turkish-gpt2-large-750m-instruct-v0.1

----------

<div style="text-align:center;">
    <img src="./model_cover.png" width="400px"/>
</div>

----------

Derived from ytu-ce-cosmos/turkish-gpt2-large, this model is a Turkish Language Model (LLM) finetuned with a dataset consisting of 35K instructions.
Due to the diverse nature of the training data, which includes websites, books, and other text sources, this model can exhibit biases and generate wrong answers. Users should be aware of these biases and use the model responsibly.

## Quickstart

```python
import torch
from transformers import AutoTokenizer, GPT2LMHeadModel
from transformers import pipeline

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

model = GPT2LMHeadModel.from_pretrained("ytu-ce-cosmos/turkish-gpt2-large-750m-instruct-v0.1").to(device)

tokenizer = AutoTokenizer.from_pretrained("ytu-ce-cosmos/turkish-gpt2-large-750m-instruct-v0.1")

text_generator = pipeline('text-generation', model=model, tokenizer=tokenizer, max_new_tokens=256)

def get_model_response(instruction):
    instruction_prompt = f"### Kullanıcı:\n{instruction}\n### Asistan:\n"
    result = text_generator(instruction_prompt)
    generated_response = result[0]['generated_text']
    return generated_response[len(instruction_prompt):]

print(get_model_response("Evde egzersiz yapmanın avantajlarını açıkla."))
"""
Evde egzersiz yapmak, gelişmiş fiziksel ve zihinsel sağlık için harika bir yoldur. Düzenli egzersizin, artan enerji seviyeleri, gelişmiş kas gücü ve esnekliği, gelişmiş uyku kalitesi ve daha iyi genel esenlik dahil olmak üzere birçok faydası vardır. Evde egzersiz yapmak ayrıca stresi azaltmaya, kas gücünü artırmaya ve genel sağlığı iyileştirmeye yardımcı olabilir.
"""
```


----------

### Training Details

- We've meticulously fine-tuned this model with a 35,000-instruction Turkish dataset to enhance its precision and adaptability.

- By employing LoRA (Low-Rank Adaptation), we have successfully propelled this model to the pinnacle of its performance capabilities.
- **LoRA** Config:
    * rank = 256
    * lora_alpha = 512
    * lora_dropout = 0.05
    * bias="none"
    * task_type="CAUSAL_LM"

- In addition to monitoring loss, we successfully integrated Rouge calculations into our system's evaluation metrics.
- One of the innovative techniques we adopted involved employing a model to cleanse our data.

*Further details will be provided in the forthcoming paper.*

----------

### Model Description
- **Developed by:** cosmos-ytuce
- **Finetuned from model :** `ytu-ce-cosmos/turkish-gpt2-large`

# Acknowledgments
- Thanks to the generous support from the Hugging Face team, it is possible to download models from their S3 storage 🤗

----------
### Citation
Paper coming soon 😊

----------

### Framework versions

- PEFT 0.9.0