Vision MLP GELU backward
Backward pass for Vision MLP with GELU activation. Computes gradients for input hidden state, fc1 weight/bias, and fc2 weight/bias through the chain: fc2 -> GELU -> fc1.
Reported evidence · last observed 2026-08-24. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 16 cases · mean latencyhidden_size = 1280 · intermediate_size = 358416 cases
Not measured on H100 for this workload. Challenges →
171.9µs70% SOL1.00×Reported · license unknown · no source2026-08-24
Measured exactly what you asked. Reported by source; not independently reproduced.
281.6µs67% SOL1.13×Reported · license unknown · no source2026-07-15
Measured exactly what you asked. Reported by source; not independently reproduced.
382.0µs66% SOL1.14×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
487.2µs65% SOL1.21×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
587.2µs65% SOL1.21×Reported · license unknown · no source2026-07-20
Measured exactly what you asked. Reported by source; not independently reproduced.
689.9µs64% SOL1.25×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
790.8µs64% SOL1.26×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
895.1µs62% SOL1.32×Reported · license unknown · no source2026-07-12
Measured exactly what you asked. Reported by source; not independently reproduced.
9103.0µs60% SOL1.43×Reported · license unknown · no source2026-08-14
Measured exactly what you asked. Reported by source; not independently reproduced.
10105.8µs59% SOL1.47×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
11108.9µs59% SOL1.51×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
12114.0µs57% SOL1.59×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
13122.2µs56% SOL1.70×Reported · license unknown · no source2026-08-24
Measured exactly what you asked. Reported by source; not independently reproduced.
14126.6µs54% SOL1.76×Reported · license unknown · no source2026-08-07
Measured exactly what you asked. Reported by source; not independently reproduced.
15132.4µs53% SOL1.84×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
16138.0µs52% SOL1.92×Reported · license unknown · no source2026-08-10
Measured exactly what you asked. Reported by source; not independently reproduced.
17144.5µs51% SOL2.01×Reported · license unknown · no source2026-08-05
Measured exactly what you asked. Reported by source; not independently reproduced.
18144.7µs51% SOL2.01×Reported · license unknown · no source2026-08-11
Measured exactly what you asked. Reported by source; not independently reproduced.
19181.0µs45% SOL2.52×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
20191.4µs43% SOL2.66×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
21329.9µs30% SOL4.59×Reported · license unknown · no source2026-08-13
Measured exactly what you asked. Reported by source; not independently reproduced.
221.24ms10% SOL17.20×Reported · license unknown · no source2026-08-02
Measured exactly what you asked. Reported by source; not independently reproduced.