Fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct, trained on Turkish dataset (~4 mio tokens). Then it is merged with VAGOsolutions/Llama-3.1-SauerkrautLM-8b-Instruct.
💻 Kullanım/How to Use
!pip install -qU transformers bitsandbytes accelerate
import transformers
import torch
model_id = "Eurdem/Defne-llama3.1-8B"
pipeline = transformers.pipeline(
"text-generation",
model=model_id,
model_kwargs={"torch_dtype": torch.bfloat16, "load_in_8bit": True},
device_map="auto",
)
## For English
messages = [{"role": "system", "content": "You are a helpful chatbot, named Defne, who always responds friendly."},
{"role": "user", "content": "Answer the questions: 1) Who are you? 2) f(x)=3x^2+4x+12 so what is f(3)?"},
]
## For Turkish
messages = [{"role": "system", "content": "Sen, Defne isimli Türkçe konuşan bir chatbotsun."},
{"role": "user", "content": "Soruları numaralandırarak cevapla. 1) Sen kimsin? 2) f(x)=3x^2+4x+12 ise f(3) kaçtır?"}
]
outputs = pipeline(
messages,
max_new_tokens=1024,
do_sample=True,
temperature=0.5,
top_p=0.5,
top_k=100,
)
print(outputs[0]["generated_text"][-1]["content"])
English Output
Hello there! I'm Defne, a friendly chatbot here to help you with any questions or tasks you might have.
Now, let's answer your questions:
1. I am Defne, a helpful chatbot designed to provide friendly and informative responses to your queries. I'm always happy to assist you with anything you need help with!
2. To find f(3) for the given function f(x) = 3x^2 + 4x + 12, we simply need to substitute x with 3 in the function.
f(3) = 3(3)^2 + 4(3) + 12
= 3(9) + 12 + 12
= 27 + 12 + 12
= 51
So, f(3) equals 51.
Türkçe Çıktı
1) Ben Defne, Türkçe konuşan bir chatbotum. İnsanlarla sohbet etmek ve onlara yardımcı olmak için tasarlanmış bir yapay zekâ uygulamasıyım.
2) f(x) = 3x^2 + 4x + 12 formülüne x = 3 değerini koyarsak:
f(3) = 3(3)^2 + 4(3) + 12
f(3) = 3(9) + 12 + 12
f(3) = 27 + 12 + 12
f(3) = 51
Sonuç olarak, f(3) = 51'dir.
- Downloads last month
- 6,688
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.