gpt-python-modern-114m

A ~114M parameter LLaMA/Mistral-style decoder-only transformer (RMSNorm, RoPE, SwiGLU, Grouped Query Attention), trained from scratch on devoppro/Python.

• Layers: 12, Query heads: 12, KV heads: 4
• Embedding dim: 768
• Context length: 512
• Trained to step 0, val loss 10.3440
• Tokenizer: GPT-2 BPE (tiktoken 'gpt2')