Checks / edit
Weight tying lost in the edit
Check C02. Compares a change with what it was changed from, in
POST /api/v1/plan and in the pull request comment. Milliseconds, before any GPU is billed.
warn
edit
C02
| Trigger | The base ties the output head to the input embedding (tie_word_embeddings: true, or a head with tied) and the changed model sets it false, drops the key, or keeps the tie on a head whose shape no longer matches the embedding. |
|---|---|
| Why | An untied head is a second vocab × hidden matrix that drifts from the embedding from the first step. When the flag is dropped on save, the reload decides by class default and comes back untied with no error. |
| Seen in | transformers#38160 (tie_word_embeddings not saved on a customised model), peft#2944. |
Why it is not a lint you can ignore
This edit does not fail at all. It loads, it runs, and the save succeeds, so a
coding agent that fixes what errors has nothing to fix. The first sign is a training run that
learned the wrong thing. 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
50 structural checks: 6 guardrail gates, 37 architecture advisor rules and 7 edit checks. See the full catalogue.
← C01 Vocabulary resized, a token-indexed tensor was not · C03 Attention config keys that did not move together →