YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
My From-Scratch MLX Transformer
A small transformer language model built from scratch using MLX, trained on a tiny text corpus as part of a hands-on learning course.
Architecture
- Embedding dimension: 64
- Attention heads: 4
- Transformer blocks: 4
- FFN dimension: 256
- Vocabulary: GPT-2 tokenizer (50257 tokens)
- Max sequence length: 128
Usage
Load weights with mlx.core.load and reconstruct the architecture
using the values in config.json.
Intended Use
Educational demonstration only. Not suitable for production use.
- Downloads last month
- 3
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support