busratuter/metal-rock Model

---

English Description

This model is a fine-tuned version of Meta's Llama 3.1 8B model, trained on a custom Turkish dataset focused on rock and metal music. The model's purpose is to answer questions about these music genres, explain the similarities between bands, and provide general knowledge.

The training was accelerated and optimized using the Unsloth library.

• Developer: busratuter
• License: apache-2.0
• Finetuned from model: unsloth/meta-llama-3.1-8b-unsloth-bnb-4bit

Model Description

This model is designed as an assistant for users who want to master the world of rock and metal music, learn about the musical styles of various bands, and discover similarities between them. It contains information about bands like AC/DC, Metallica, Queen, and Green Day, as well as many different subgenres.

How to Use (with GGUF)

The GGUF version of this model can be run efficiently on your local machine using llama.cpp-based applications (like Jan, LM Studio, Open WebUI, etc.).

Below is an up-to-date and correct example of how to use it with the llama-cpp-python library:

# Install the required libraries
!pip install llama-cpp-python huggingface_hub -q

from huggingface_hub import hf_hub_download
from llama_cpp import Llama

# Download the model from the Hugging Face Hub
model_name = "busratuter/metal-rock"
model_file = "unsloth.Q4_K_M.gguf" # The name of the GGUF file you want to use
model_path = hf_hub_download(repo_id=model_name, filename=model_file)

# Load the model, specifying the Llama 3 chat format
llm = Llama(
    model_path=model_path,
    n_ctx=4096,          # Maximum context window to use
    n_threads=8,         # Number of CPU cores to use
    n_gpu_layers=-1,     # Number of layers to offload to GPU (-1 for all)
    chat_format="llama-3" # This line automatically sets the prompt format
)

# Create a message list suitable for the Llama 3.1 Chat format
messages = [
    {
        "role": "system",
        "content": "You are an expert assistant on rock and metal music. Answer questions with your knowledge in this field, providing clear and informative responses.",
    },
    {
        "role": "user",
        "content": "What are the main musical differences between Iron Maiden and Metallica?",
    },
]

# Run the model in chat mode
output = llm.create_chat_completion(
      messages = messages,
      max_tokens = 512, # Maximum response length
)

# Print the model's response
print(output["choices"][0]["message"]["content"])