onenight's starred repositories
gpt_academic
为GPT/GLM等LLM大语言模型提供实用化交互接口,特别优化论文阅读/润色/写作体验,模块化设计,支持自定义快捷按钮&函数插件,支持Python和C++等项目剖析&自译解功能,PDF/LaTex论文翻译&总结功能,支持并行问询多种LLM模型,支持chatglm3等本地模型。接入通义千问, deepseekcoder, 讯飞星火, 文心一言, llama2, rwkv, claude2, moss等。
LLaMA-Factory
Efficiently Fine-Tune 100+ LLMs in WebUI (ACL 2024)
tuning_playbook
A playbook for systematically maximizing the performance of deep learning models.
RWKV-LM
RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
awesome-knowledge-distillation
Awesome Knowledge Distillation
hugging-llm
HuggingLLM, Hugging Future.
TextBrewer
A PyTorch-based knowledge distillation toolkit for natural language processing
lm-human-preferences
Code for the paper Fine-Tuning Language Models from Human Preferences
awesome_LLMs_interview_notes
LLMs interview notes and answers:该仓库主要记录大模型(LLMs)算法工程师相关的面试题和参考答案
data-centric-AI
A curated, but incomplete, list of data-centric AI resources.
RAG-Retrieval
Unify Efficient Fine-tuning of RAG Retrieval, including Embedding, ColBERT,Cross Encoder
legal-ml-datasets
A collection of datasets and tasks for legal machine learning
awesome-llm-understanding-mechanism
awesome papers in LLM interpretability
self-speculative-decoding
Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**
Pruning-LLMs
The framework to prune LLMs to any size and any config.
FudanNLP_Begginer
这是复旦NLP实验室五个任务的代码仓库,每个代码以ipynb形式展开,可以通过colab在线运行代码
hibrids_summ
Code for ACL 2022 paper "HIBRIDS: Attention with Hierarchical Biases for Structure-aware Long Document Summarization".