bkitano / llama-from-scratch
Llama from scratch, or How to implement a paper without crying
☆515Updated 5 months ago
Related projects ⓘ
Alternatives and complementary repositories for llama-from-scratch
- ☆679Updated last month
- Deep learning for dummies. All the practical details and useful utilities that go into working with real models.☆706Updated last month
- LLM Workshop by Sourab Mangrulkar☆340Updated 4 months ago
- What would you do with 1000 H100s...☆894Updated 9 months ago
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,143Updated 3 weeks ago
- From scratch implementation of a sparse mixture of experts language model inspired by Andrej Karpathy's makemore :)☆593Updated last week
- Minimalistic large language model 3D-parallelism training☆1,227Updated this week
- An open collection of methodologies to help with successful training of large language models.☆459Updated 8 months ago
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.☆642Updated last month
- Best practices for distilling large language models.☆392Updated 9 months ago
- Fine-tune mistral-7B on 3090s, a100s, h100s☆702Updated last year
- Puzzles for learning Triton☆1,068Updated last month
- A minimum example of aligning language models with RLHF similar to ChatGPT☆213Updated last year
- Serving multiple LoRA finetuned LLM as one☆979Updated 6 months ago
- ☆1,263Updated this week
- Official repository for ORPO☆420Updated 5 months ago
- Extend existing LLMs way beyond the original training length with constant memory usage, without retraining☆667Updated 6 months ago
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆661Updated this week
- A bibliography and survey of the papers surrounding o1☆577Updated this week
- Building blocks for foundation models.☆386Updated 10 months ago
- A family of open-sourced Mixture-of-Experts (MoE) Large Language Models☆1,385Updated 8 months ago
- Flash Attention in ~100 lines of CUDA (forward pass only)☆609Updated 7 months ago
- LoRA and DoRA from Scratch Implementations☆188Updated 8 months ago
- Scalable toolkit for efficient model alignment☆611Updated this week
- Official implementation of Half-Quadratic Quantization (HQQ)☆698Updated last week
- ☆465Updated 2 months ago
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆479Updated 2 weeks ago
- batched loras☆336Updated last year
- LLaMA 2 implemented from scratch in PyTorch☆250Updated last year