EleutherAI / DeeperSpeed
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
☆165Updated 6 months ago
Related projects ⓘ
Alternatives and complementary repositories for DeeperSpeed
- Used for adaptive human in the loop evaluation of language and embedding models.☆303Updated last year
- ☆64Updated 2 years ago
- ☆128Updated 2 years ago
- Swarm training framework using Haiku + JAX + Ray for layer parallel transformer language models on unreliable, heterogeneous nodes☆239Updated last year
- Inference code for LLaMA models in JAX☆112Updated 5 months ago
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated last year
- Implementation of the specific Transformer architecture from PaLM - Scaling Language Modeling with Pathways - in Jax (Equinox framework)☆185Updated 2 years ago
- Experiments with generating opensource language model assistants☆97Updated last year
- Python library which enables complex compositions of language models such as scratchpads, chain of thought, tool use, selection-inference…☆195Updated 4 months ago
- Train very large language models in Jax.☆195Updated last year
- ☆86Updated 2 years ago
- Blazing fast training of 🤗 Transformers on Graphcore IPUs☆81Updated 7 months ago
- OSLO: Open Source for Large-scale Optimization☆174Updated last year
- Pipeline for pulling and processing online language model pretraining data from the web☆174Updated last year
- ☆76Updated 11 months ago
- JAX implementation of the Llama 2 model☆210Updated 9 months ago
- RWKV-v2-RNN trained on the Pile. See https://github.com/BlinkDL/RWKV-LM for details.☆66Updated 2 years ago
- some common Huggingface transformers in maximal update parametrization (µP)☆76Updated 2 years ago
- Code repository for the c-BTM paper☆105Updated last year
- Techniques used to run BLOOM at inference in parallel☆37Updated 2 years ago
- The GeoV model is a large langauge model designed by Georges Harik and uses Rotary Positional Embeddings with Relative distances (RoPER).…☆122Updated last year
- Open Instruction Generalist is an assistant trained on massive synthetic instructions to perform many millions of tasks☆206Updated 9 months ago
- Minimal code to train a Large Language Model (LLM).☆164Updated 2 years ago
- ☆333Updated 6 months ago
- ☆147Updated 3 years ago
- Simple Annotated implementation of GPT-NeoX in PyTorch☆111Updated 2 years ago
- A Multilingual Dataset for Parsing Realistic Task-Oriented Dialogs☆113Updated last year
- Scalable PaLM implementation of PyTorch☆192Updated last year
- Multipack distributed sampler for fast padding-free training of LLMs☆175Updated 3 months ago