Checks / ordering
Back-to-back activations
Check R19. Runs in the editor as you build, in CI through the
GitHub Action, and over the wire at
POST /api/v1/check. Milliseconds, before any GPU is billed.
warn
ordering
R19
| Trigger | An activation node (relu/gelu/silu/sigmoid/tanh/softmax/…) connects directly into another activation node with no Linear/Conv/Norm between them. |
|---|---|
| Why | Two activations in a row add compute but no expressivity (same-type is a no-op duplicate). A common slip is ReLU → Softmax, which clips logits to ≥ 0 and distorts the output distribution. |
| Source | Composition of pointwise non-linearities adds no representational power without an intervening affine map (universal-approximation premise). |
Why it is not a lint you can ignore
A structural mistake does not fail at review time and it does not fail at import
time. It fails when the module is constructed on the training node, after the job was queued and
the dataset was downloaded. That is why this runs before the spend and not after it.
Run this check on your own model
Free, no account needed
Every check
41 structural checks: 6 guardrail gates and 35 architecture advisor rules. See the full catalogue.
← R18 Conv feeds Linear with no flatten / pool · R20 Linear → Linear with no activation →