Downloads · 30 days
0
chessPT
philipp-zettl/chessPT
chessPT is a text generation model from philipp-zettl. Use it when you need the model to write or continue text. The card lists the license as cc0-1.0.
A pretrained Decoder only transformer model for chess move prediction with only ~14M parameters.
Access
Public
Updated Apr 26, 2026
Repo size
315 MB
Likes
7
Public
Repo makeup
Click a slice to open those files.
.pth158 MB · 50%
At a glance
- Task
- Text Generation
- License
- cc0-1.0
- Access
- Public
- Created
- Oct 1, 2024
- Updated
- Apr 26, 2026
- SHA
- e016720d
Try a prompt
Datasets
Spaces
Model card
From the Hugging Face model README
Model card for chessPT
A pretrained Decoder only transformer model for chess move prediction with only ~14M parameters.
Intended use
Predict new moves in a chess game based on PGN tokens.
Implementation
The model implementation is based on Andrej Karpathy's nanoGPT following the webseries "Zero to Hero" on youtube.
Training
You can find the training script in the repositories files under train.py.
This also contains the used parameters
context_size = 256
batch_size = 128
max_iters = 30_000
learning_rate = 3e-5
eval_interval = 100
eval_iters = 20
n_embed = 384
n_layer = 6
n_head = 6
dropout = 0.2
- Task
- Text Generation
- License
- cc0-1.0
- Created
- Oct 1, 2024
- Updated
- Apr 26, 2026