Expand description
DeepSeek-V4.1 full tiny-graph forward (Block.forward, Transformer.forward): embed,
expand to hc_mult copies, the one-hot initial pre-mix, engram before layers 1 and 4, six
blocks with the delayed mixes, collapse, final RMSNorm, f32 logits. Every component is the
CPU reference already checked on its own; this file wires them in the reference’s order and
carries every per-layer cache and the shared slots across prefill and decode, so the golden’s
sampled captures for layers 1 to 5 and the logits become checkable.
Structs§
Functions§
- forward
Transformer.forwardfor one chunk ([seqlen]ids atstart_pos), text only.