Instructions to use naxautify/gpt2-4k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use naxautify/gpt2-4k with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="naxautify/gpt2-4k")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("naxautify/gpt2-4k") model = AutoModelForCausalLM.from_pretrained("naxautify/gpt2-4k") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use naxautify/gpt2-4k with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "naxautify/gpt2-4k" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "naxautify/gpt2-4k", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/naxautify/gpt2-4k
- SGLang
How to use naxautify/gpt2-4k 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 "naxautify/gpt2-4k" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "naxautify/gpt2-4k", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "naxautify/gpt2-4k" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "naxautify/gpt2-4k", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use naxautify/gpt2-4k with Docker Model Runner:
docker model run hf.co/naxautify/gpt2-4k
How did you train / FT this?
#2
by venketh - opened
- Fine-tune GPT2 or train from scratch? (AFAICT it's challenging to fine-tune w/ different n_ctx than a base model)
- What trainer?
- I'm seeing the weights tensor have a dim of [2048,768] rather than the expected [4096,768] for "gpt2-4k"; is this a -2k context gpt2?
- Should n_ctx/n_positions in config.json be updated?