pierreandLetta Code a2ff07234c fix(linear): add bias once, not once per SIMD lane ...

acc was seeded with @splat(bias) and reduced with .Add, so every
linear output carried 7 extra copies of its bias. conv2d was unaffected
(bias via memset), which is why the trunk matched zig-solver to 5e-6
while fc/head activations diverged — the 81% labels regression.

Also adds a scalar tail for in_n < 8 (pose_fc has in_n=4) and a
known-answer selftest (test_linear).

👾 Generated with [Letta Code](https://letta.com)

Co-Authored-By: Letta Code <noreply@letta.com>

2026-09-29 22:08:45 +03:00

..

stripsolver.zig

crucible v0.1: CPU inference library for PyTorch .pt state dicts, in pure Zig

2026-09-29 18:41:46 +03:00

test_linear.zig

fix(linear): add bias once, not once per SIMD lane

2026-09-29 22:08:45 +03:00