tysam-code / hlb-gpt
Minimalistic, extremely fast, and hackable researcher's toolbench for GPT models in 307 lines of code. Reaches <3.8 validation loss on wikitext-103 on a single A100 in <100 seconds. Scales to larger models with one parameter change (feature currently in alpha).
☆343Updated 7 months ago
Alternatives and similar repositories for hlb-gpt:
Users that are interested in hlb-gpt are comparing it to the libraries listed below
- a small code base for training large models☆288Updated 2 months ago
- Puzzles for exploring transformers☆334Updated last year
- Ungreedy subword tokenizer and vocabulary trainer for Python, Go & Javascript☆571Updated 8 months ago
- Fully fine-tune large models like Mistral, Llama-2-13B, or Qwen-14B completely for free☆230Updated 4 months ago
- ☆92Updated last year
- An interactive exploration of Transformer programming.☆259Updated last year
- A puzzle to learn about prompting☆124Updated last year
- ☆412Updated last year
- ☆212Updated 7 months ago
- Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax☆554Updated this week
- Language Modeling with the H3 State Space Model☆515Updated last year
- A really tiny autograd engine☆90Updated 11 months ago
- batched loras☆338Updated last year
- JAX implementation of the Llama 2 model☆216Updated last year
- Fine-tune mistral-7B on 3090s, a100s, h100s☆706Updated last year
- ☆143Updated last year
- Fast bare-bones BPE for modern tokenizer training☆148Updated 4 months ago
- A Jax-based library for designing and training transformer models from scratch.☆282Updated 6 months ago
- ☆418Updated 4 months ago
- Extract full next-token probabilities via language model APIs☆231Updated last year
- The history files when recording human interaction while solving ARC tasks☆97Updated this week
- Helpers and such for working with Lambda Cloud☆52Updated last year
- The repository for the code of the UltraFastBERT paper☆517Updated 11 months ago
- seqax = sequence modeling + JAX☆148Updated 2 weeks ago
- A tool to analyze and debug neural networks in pytorch. Use a GUI to traverse the computation graph and view the data from many different…☆281Updated 3 months ago
- Website for hosting the Open Foundation Models Cheat Sheet.☆263Updated this week
- Annotated version of the Mamba paper☆474Updated last year
- Visualizing the internal board state of a GPT trained on chess PGN strings, and performing interventions on its internal board state and …☆201Updated 3 months ago
- ☆511Updated 6 months ago
- A crude RLHF layer on top of nanoGPT with Gumbel-Softmax trick☆289Updated last year