GPAS scales down intermediate activations while preserving backward gradients, reducing activation variance growth in Pre-LN transformers and improving pretraining convergence and downstream performance.
Adaptive input representations for neural language modeling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
method 1polarities
use method 1representative citing papers
citing papers explorer
-
GPAS: Accelerating Convergence of LLM Pretraining via Gradient-Preserving Activation Scaling
GPAS scales down intermediate activations while preserving backward gradients, reducing activation variance growth in Pre-LN transformers and improving pretraining convergence and downstream performance.