LightningConversation

llmsearch

Find better generation parameters for your LLM

ProteinDT

plain-transformers

Transformer models implementation for training from scratch.

multi_token

Embed arbitrary modalities (images, audio, documents, etc) into large language models.

NoticIA

We present NoticIA, a dataset consisting of 850 Spanish news articles featuring prominent clickba...

archives

python-packages-for-data-geeks

A curated list of useful Python packages for data geeks

GPT2

An implementation of training for GPT2, supports TPUs

instructrl

Instruction Following Agents with Multimodal Transforemrs

TristouNet

TristouNet: Triplet Loss for Speaker Turn Embedding

summarize_from_feedback_details

chain-of-hindsight

Chain-of-Hindsight, A Scalable RLHF Method

fatigue-detection

本项目基于PaddleDetection目标检测开发套件，选取1.3M超轻量PPYOLO tiny进行项目开发，并部署于windows端。

TUPE

Transformer with Untied Positional Encoding (TUPE). Code of paper "Rethinking Positional Encoding...

pnlp

NLP预/后处理工具。