tokenization: subword -> token-free BLT (raw bytes + patches)

Encoder-decoder Transformer built from basic PyTorch operations (no nn.Transformer or nn.MultiheadAttention), trained to decrypt binary cipher text into English plaintext.

One of five configurations in a controlled ablation. Each of C2-C5 changes exactly one component from the C1 base; every other hyperparameter is identical across all five runs.

d_model 256 | 4 encoder layers | 4 decoder layers
8 heads | d_ff 1024 | dropout 0.1 | pre-norm
AdamW lr 0.0003 | warmup 1000 | 15000 steps | batch 64 | seed 42
fp32 (no AMP)

Checkpoints are plain PyTorch state dicts. See the accompanying code repository for the model definition; torch.load(...)["model"] loads into build_model(get_config("c5"), ...).