Dorna Llama3 8B Instruct

🚀 Dorna-Llama3-8B-Instruct

The Dorna models are decoder-only models, fine - tuned on Persian data by Part AI. This release offers an 8B instruct model, built upon Meta Llama 3 Instruct.

🚀 Quick Start

You can run conversational inference using the Transformers Auto classes with the generate() function.

✨ Features

• Family of decoder - only models.
• Specifically trained/fine - tuned on Persian data.
• Built using the Meta Llama 3 Instruct model.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

import torch
import transformers
from transformers import AutoTokenizer, AutoModelForCausalLM


tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(
    model_path,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

messages = [
    {"role": "system",
     "content": "You are a helpful Persian assistant. Please answer questions in the asked language."},
    {"role": "user", "content": "کاغذ A4 بزرگ تر است یا A5؟"},
]

input_ids = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

terminators = [
    tokenizer.eos_token_id,
    tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = model.generate(
    input_ids,
    max_new_tokens=256,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.6,
    top_p=0.9,
)
response = outputs[0][input_ids.shape[-1]:]
print(tokenizer.decode(response, skip_special_tokens=True))

You can also use the notebook below to test the model in Google Colab.

📚 Documentation

Evaluation

This model is evaluated on various tasks, including Boolean Questions, Code Generation, etc. Both human evaluation and automatic evaluation (with GPT - 4 as the judge) are performed.

Human Evaluation

Overall human evaluation results:

Category - based human evaluation results:

Automatic Evaluation

🔧 Technical Details

No technical details are provided in the original document, so this section is skipped.

📄 License

The license of this model is llama3.

Contact us

If you have any questions regarding this model, you can reach us via the community on Hugging Face.