Text Generation
Transformers
Safetensors
English
Persian
llama
LLM
llama-3
PartAI
conversational
text-generation-inference
Instructions to use PartAI/Dorna-Llama3-8B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use PartAI/Dorna-Llama3-8B-Instruct with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="PartAI/Dorna-Llama3-8B-Instruct") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("PartAI/Dorna-Llama3-8B-Instruct") model = AutoModelForCausalLM.from_pretrained("PartAI/Dorna-Llama3-8B-Instruct") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use PartAI/Dorna-Llama3-8B-Instruct with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "PartAI/Dorna-Llama3-8B-Instruct" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "PartAI/Dorna-Llama3-8B-Instruct", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/PartAI/Dorna-Llama3-8B-Instruct
- SGLang
How to use PartAI/Dorna-Llama3-8B-Instruct with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "PartAI/Dorna-Llama3-8B-Instruct" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "PartAI/Dorna-Llama3-8B-Instruct", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "PartAI/Dorna-Llama3-8B-Instruct" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "PartAI/Dorna-Llama3-8B-Instruct", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use PartAI/Dorna-Llama3-8B-Instruct with Docker Model Runner:
docker model run hf.co/PartAI/Dorna-Llama3-8B-Instruct
fine tuning
#5
by Artin2009 - opened
سلام وقت بخیر
چطوری میشه این مدل رو فاین تیون کرد ؟
سلام، وقت شما هم بخیر.
برای تنظیم دقیق مدلها با توجه به منابع (داده و سختافزار) موجود، رویکردهای مختلفی وجود دارد که بدون اطلاع دقیق از منابع شما، نمیتوانم پیشنهاد مشخصی بدم. اما به طور کلی، میتوانید روشهای زیر را مطالعه کرده و با توجه به منابع خود، مناسبترین روش را انتخاب کنید:
- SFT
- LoRA
- QLoRA
- DPO
- PPO
- ORPO
- RLHF
یکی از بهترین ابزارها هم transformers هست.
خیلی ممنون. ایا امکان فاین تیونیگ با روش REFT هست ؟
و اینکه شما برای روش SFT چه چیزی پیشنهاد میکنید ؟
MiladMola changed discussion status to closed