tensor-fusion/GPT-Haskell
A pure Haskell implementation of a decoder-only transformer (GPT)
This project helps researchers and students understand how Large Language Models (LLMs) work by providing a simplified version of the GPT-2 architecture. It takes GPT-2 model weights and tokenizer configurations as input and allows you to explore the internal workings of the decoder-only transformer and text generation. It's designed for those learning about or teaching deep learning and natural language processing.
No commits in the last 6 months.
Use this if you are a computer science student, researcher, or educator who wants to study the core mechanics of a GPT-like model implemented in a functional programming language like Haskell.
Not ideal if you're looking for a production-ready tool to build or deploy large-scale language models, or if you need to perform advanced natural language processing tasks.
Stars
21
Forks
1
Language
Haskell
License
—
Category
Last pushed
Jun 22, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/tensor-fusion/GPT-Haskell"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
LowinLi/transformers-stream-generator
This is a text generation method which returns a generator, streaming out each token in...
ystemsrx/mini-nanoGPT
One-click training of your own GPT. Training a GPT has never been easier for beginners. /...
jaymody/picoGPT
An unnecessarily tiny implementation of GPT-2 in NumPy.
kyegomez/AttentionGrid
A network of attention mechanisms at your fingertips. Unleash the potential of attention...
kamalkraj/minGPT-TF
A minimal TF2 re-implementation of the OpenAI GPT training