LLaMA 33b finetuned on wikitext_document_level with a linear ROPE scaling of 8, for a 16k token context length.
This is a merged version of llama33b-16k-qlora.
Note that this is not an instruct model - this is base LLaMA with an extended sequence length.
- Downloads last month
- 4
Install from pip and serve model
# Install vLLM from pip: pip install vllm# Start the vLLM server: vllm serve "chargoddard/llama33b-16k"# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "chargoddard/llama33b-16k", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'