v2 card: 12288-vocab retrain, 56,902,144 params, val 1.3837, measured evals e217b62 verified Compactbot commited on 2 days ago
Add weights (54,804,992 params, 99 tensors, float32) 1d462bb verified Compactbot commited on 4 days ago
Add architecture config (D512/L16/H8/FFN2048/SEQ512/VOCAB8192, weight-tied) f94f9c1 verified Compactbot commited on 4 days ago
Add model card: 54.8M from-scratch TinyStories GPT, verified params + held-out PPL 5.24 8af5ac4 verified Compactbot commited on 4 days ago