QWEN3 4B · FINETOME

Evidence record

An experiment record that separates a working pipeline from improved model quality.

Verified30 / 30training steps completed
Verified0.8245final training loss
Verifiedadapter save and reload
Observed9.62 / 15.99 GiBobserved snapshot; not peak

VERDICT

Pipeline passed. Quality gain was not proven.

There was no evaluation split, and the base model was stronger in all three fixed quality comparisons. This run proves a smoke-test pipeline, not a model improvement.

Turkish explanationBase more coherentFine-tuned repetition/language mixing
Conditional JSONBase more validFine-tuned incomplete/loose
Missing informationBase more cautiousFine-tuned made more assumptions

Unknowns

  • Real peak VRAM was not measured.
  • There was no independent validation/test split.
  • There was no fixed seed or repeated run.
  • Statistical confidence of the quality delta is unknown.