llm-playground

A sandbox environment to experiment with large language models and various NLP tasks.

Notes

• Hardware: NVIDIA-SMI 550.107.02 Driver Version: 550.107.02 CUDA Version: 12.4
• Python: Python =3.8
• Use the huggingface-cli to download a specific model. For example: huggingface-cli download model <repo_id>
• Install cudatoolkit in conda: conda install -c conda-forge cudatoolkit-dev -y

references

llama-3.2-90b-vision-instruct

nlp-course

promptingguide

natural-language-processing

generative-ai-engineering-with-llms

mlops-machine-learning-duke

large-language-model-operations

speech-and-language-processing

sentence-transformer

huggingface-transformers

huggingface-repo-datasets

huggingface-repo-tokenizers

huggingface-repo-accelerate

huggingface-repo-evaluate

huggingface-repo-autotrain-advanced

huggingface-models-hub

huggingface-learn

huggingface-autotrain

huggingface-hugging-quants

sft-llama3

sft_trainer

huggingface-llama-recipes

alignment-handbook

faiss

openai-chatgpt-api-pricing-calculator