minimal pytorch implementation of bm25 (with sparse tensors)
MIT License
generative language model training on top of the JAX and Huggingface 🤗
Home of StarCoder2!
Auto summarization from BPE tokenization
Code for paper Fine-tune BERT for Extractive Summarization
NAACL 2021 - Progressive Generation of Long Text
ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings - NeurIPS 20...
Home of StarCoder: fine-tuning & inference!
keras implement of transformers for humans
Explore training for quantized models
Tensorflow implementation of contextualized word representations from bi-directional language models
Document Search Engine project with TF-IDF abd Google universal sentence encoder model
Ongoing research training transformer language models at scale, including: BERT & GPT-2
Temporary remove unused tokens during training to save ram and speed.
Some preliminary explorations of Mamba's context scaling.