tysam-code / hlb-gptLinks
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).
☆352Updated last year
Alternatives and similar repositories for hlb-gpt
Users that are interested in hlb-gpt are comparing it to the libraries listed below
Sorting:
- ☆94Updated 2 years ago
- Fast bare-bones BPE for modern tokenizer training☆171Updated 5 months ago
- Simple Transformer in Jax☆139Updated last year
- ☆415Updated 2 years ago
- JAX implementation of the Llama 2 model☆216Updated last year
- Ungreedy subword tokenizer and vocabulary trainer for Python, Go & Javascript☆606Updated last year
- a small code base for training large models☆315Updated 7 months ago
- The repository for the code of the UltraFastBERT paper☆520Updated last year
- Language Modeling with the H3 State Space Model☆519Updated 2 years ago
- ☆144Updated 2 years ago
- ☆314Updated last year
- Fully fine-tune large models like Mistral, Llama-2-13B, or Qwen-14B completely for free☆232Updated last year
- A repository for log-time feedforward networks☆223Updated last year
- git extension for {collaborative, communal, continual} model development☆216Updated last year
- Fine-tune mistral-7B on 3090s, a100s, h100s☆717Updated 2 years ago
- Puzzles for exploring transformers☆380Updated 2 years ago
- A crude RLHF layer on top of nanoGPT with Gumbel-Softmax trick☆293Updated 2 years ago
- A puzzle to learn about prompting☆135Updated 2 years ago
- An interactive exploration of Transformer programming.☆270Updated 2 years ago
- batched loras☆347Updated 2 years ago
- Extract full next-token probabilities via language model APIs☆248Updated last year
- Full finetuning of large language models without large memory requirements☆94Updated 2 months ago
- NeurIPS Large Language Model Efficiency Challenge: 1 LLM + 1GPU + 1Day☆258Updated 2 years ago
- Helpers and such for working with Lambda Cloud☆51Updated 2 years ago
- Visualizing the internal board state of a GPT trained on chess PGN strings, and performing interventions on its internal board state and …☆219Updated last year
- Inference code for Persimmon-8B☆412Updated 2 years ago
- A Jax-based library for building transformers, includes implementations of GPT, Gemma, LlaMa, Mixtral, Whisper, SWin, ViT and more.☆297Updated last year
- Annotated version of the Mamba paper☆491Updated last year
- Convolutions for Sequence Modeling☆904Updated last year
- Open weights language model from Google DeepMind, based on Griffin.☆654Updated 6 months ago