CoinCheung / gdGPT
Train llm (bloom, llama, baichuan2-7b, chatglm3-6b) with deepspeed pipeline mode. Faster than zero/zero++/fsdp.
☆90Updated 9 months ago
Related projects ⓘ
Alternatives and complementary repositories for gdGPT
- ☆82Updated last year
- Ongoing research training transformer language models at scale, including: BERT & GPT-2☆69Updated last year
- A prototype repo for hybrid training of pipeline parallel and distributed data parallel with comments on core code snippets. Feel free to…☆49Updated last year
- [ACL 2024] Long-Context Language Modeling with Parallel Encodings☆144Updated 5 months ago
- Finetuning LLaMA with RLHF (Reinforcement Learning with Human Feedback) based on DeepSpeed Chat☆106Updated last year
- Repository of LV-Eval Benchmark☆48Updated 2 months ago
- train llama on a single A100 80G node using 🤗 transformers and 🚀 Deepspeed Pipeline Parallelism☆207Updated last year
- A flexible and efficient training framework for large-scale alignment tasks☆206Updated this week
- Codes for the paper "∞Bench: Extending Long Context Evaluation Beyond 100K Tokens": https://arxiv.org/abs/2402.13718☆285Updated last month
- CLongEval: A Chinese Benchmark for Evaluating Long-Context Large Language Models☆38Updated 8 months ago
- A MoE impl for PyTorch, [ATC'23] SmartMoE☆57Updated last year
- [EMNLP 2024] LongAlign: A Recipe for Long Context Alignment of LLMs☆217Updated 6 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆357Updated this week
- 🐋 An unofficial implementation of Self-Alignment with Instruction Backtranslation.☆132Updated 4 months ago
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**☆138Updated 5 months ago
- ☆88Updated last month
- InsTag: A Tool for Data Analysis in LLM Supervised Fine-tuning☆218Updated last year
- [ACL'24] Superfiltering: Weak-to-Strong Data Filtering for Fast Instruction-Tuning☆125Updated 2 months ago
- code for Scaling Laws of RoPE-based Extrapolation☆70Updated last year
- Clustering and Ranking: Diversity-preserved Instruction Selection through Expert-aligned Quality Estimation☆67Updated last week
- A LLaMA1/LLaMA12 Megatron implement.☆27Updated 11 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆76Updated last month
- A repository sharing the literatures about long-context large language models, including the methodologies and the evaluation benchmarks☆252Updated 3 months ago
- ☆120Updated 7 months ago
- REST: Retrieval-Based Speculative Decoding, NAACL 2024☆176Updated last month
- ☆64Updated 3 months ago
- ☆102Updated last year
- Spec-Bench: A Comprehensive Benchmark and Unified Evaluation Platform for Speculative Decoding (ACL 2024 Findings)☆188Updated 3 weeks ago
- Dataset and evaluation script for "Evaluating Hallucinations in Chinese Large Language Models"☆109Updated 5 months ago
- NTK scaled version of ALiBi position encoding in Transformer.☆66Updated last year