Experimental small LLM project for learning and testing—outputs are often nonsensical.
-
Updated
Jun 12, 2025 - Python
#
custom-tokenizer
Here are 2 public repositories matching this topic...
A from-scratch PyTorch LLM implementing Sparse Mixture-of-Experts (MoE) with Top-2 gating. Integrates modern Llama-3 components (RMSNorm, SwiGLU, RoPE, GQA) and a custom-coded Byte-Level BPE tokenizer. Pre-trained on a curated corpus of existential & dark philosophical literature.
python pytorch transformer moe from-scratch mixture-of-experts bpe gqa llm rmsnorm swiglu existential-ai llama-3-architecture rope-embeddings custom-tokenizer
-
Updated
Jan 7, 2026 - Python
Improve this page
Add a description, image, and links to the custom-tokenizer topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the custom-tokenizer topic, visit your repo's landing page and select "manage topics."