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
- ☆92Updated last year
- ☆143Updated last year
- a small code base for training large models☆288Updated 3 months ago
- Fully fine-tune large models like Mistral, Llama-2-13B, or Qwen-14B completely for free☆230Updated 4 months ago
- Fast bare-bones BPE for modern tokenizer training☆149Updated 5 months ago
- ☆412Updated last year
- A puzzle to learn about prompting☆124Updated last year
- Comprehensive analysis of difference in performance of QLora, Lora, and Full Finetunes.☆82Updated last year
- JAX implementation of the Llama 2 model☆216Updated last year
- Ungreedy subword tokenizer and vocabulary trainer for Python, Go & Javascript☆574Updated 8 months ago
- Simple Transformer in Jax☆136Updated 9 months ago
- Puzzles for exploring transformers☆333Updated last year
- ☆214Updated 8 months ago
- Helpers and such for working with Lambda Cloud☆51Updated last year
- Extract full next-token probabilities via language model APIs☆233Updated last year
- Full finetuning of large language models without large memory requirements☆93Updated last year
- Solve puzzles. Learn CUDA.☆63Updated 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 4 months ago
- Code for paper: "QuIP: 2-Bit Quantization of Large Language Models With Guarantees"☆362Updated last year
- A crude RLHF layer on top of nanoGPT with Gumbel-Softmax trick☆288Updated last year
- ☆301Updated 9 months ago
- Understand and test language model architectures on synthetic tasks.☆185Updated 2 weeks ago
- Website for hosting the Open Foundation Models Cheat Sheet.☆264Updated last week
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆122Updated 11 months ago
- The repository for the code of the UltraFastBERT paper☆517Updated 11 months ago
- run paligemma in real time☆131Updated 10 months ago
- An interactive exploration of Transformer programming.☆261Updated last year
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)☆186Updated 9 months ago
- NeurIPS Large Language Model Efficiency Challenge: 1 LLM + 1GPU + 1Day☆255Updated last year
- Annotated version of the Mamba paper☆475Updated last year