Gahow Wang 39df0b40c1 gqa: fix kv-proj shape test param indices (embed,attn_norm precede wq) ...

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-06-18 01:38:42 +08:00

..

batched.rs

model: batched forward [B,S]

2026-06-16 00:44:25 +08:00

bf16.rs

test: bf16 test reads f32-cast logits (forward now returns bf16)

2026-06-16 14:29:24 +08:00

dropout.rs

test: flash+dropout cross-feature grad-check (Phase-2 integration)

2026-06-18 00:43:54 +08:00

flash.rs

test: flash==composed bf16 uses robust mean/p99 metric (repo convention)

2026-06-17 23:19:08 +08:00

gqa.rs

gqa: fix kv-proj shape test param indices (embed,attn_norm precede wq)

2026-06-18 01:38:42 +08:00

overfit.rs

model: tiny RoPE+RMSNorm+SwiGLU transformer + overfit test

2026-06-15 16:05:20 +08:00

parity_dump.rs

gqa: real grouped-query attention (repeat_kv op + both SDPA paths + wiring + tests)

2026-06-18 01:37:37 +08:00

parity.py

gqa: real grouped-query attention (repeat_kv op + both SDPA paths + wiring + tests)

2026-06-18 01:37:37 +08:00

recompute.rs

docs: Phase T13 — activation recompute

2026-06-17 09:45:16 +08:00