README.md

---
base_model: google/gemma-2-9b-it
license: mit
language:
- tr
---

<img src="https://huggingface.co/CerebrumTech/cere-gemma-2-9b-tr/resolve/main/cereGemma.png"
alt="CEREBRUM LLM" width="420"/>
# Cere-llm-gemma-2-9b-it Model Card

Cere-llm-gemma-2-9b-it is a finetuned version of gemma-2-9b-it. It is trained on synthetically generated and natural preference datasets.

## Model Details

### Model Description

We fine-tuned [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) 

- **Developed by:** Cerebrum Tech
- **Model type:** Causal Language Model
- **License:** gemma
- **Finetuned from model:** [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it)


## How to Get Started with the Model
```
import torch
from transformers import pipeline

model_id = "Cerebrum/cere-llm-gemma-2-ito"

generator = pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device="cuda",
)
outputs = generator([{"role": "user", "content": "Türkiye'nin başkenti neresidir?"}],
                      do_sample=False,
                      eos_token_id=[generator.tokenizer.convert_tokens_to_ids("<end_of_turn>"), generator.tokenizer.eos_token_id],
                      max_new_tokens=200)
print(outputs[0]['generated_text'])
```