Repeating 0.1% of training data 100 times degrades an 800M parameter model's performance to that of a 400M model by damaging copying mechanisms and induction heads associated with generalization.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2022 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Scaling Laws and Interpretability of Learning from Repeated Data
Repeating 0.1% of training data 100 times degrades an 800M parameter model's performance to that of a 400M model by damaging copying mechanisms and induction heads associated with generalization.