Papers › Improving Transformer Models by Reordering their Sublayers
Improving Transformer Models by Reordering their Sublayers
Ofir Press, Noah A. Smith, Omer Levy
Multilayer transformer networks consist of interleaved self-attention and feedforward sublayers. Could ordering the sublayers in a different pattern lead to better performance? We generate randomly ordered transformers and train them with the language modeling objective. We observe that some of these models are able to achieve better performance than the interleaved baseline, and that those successful variants tend to have more self-attention at the bottom and more feedforward sublayers at the top. We propose a new transformer pattern that adheres to this property, the sandwich transformer, and show that it improves perplexity on multiple word-level and character-level language modeling benchmarks, at no cost in parameters, memory, or training time. However, the sandwich reordering pattern does not guarantee performance gains across every task, as we demonstrate on machine translation models. Instead, we suggest that further exploration of task-specific sublayer reorderings is needed in order to unlock additional gains.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Language Modelling | WikiText-103 | Sandwich Transformer | Number of params | 247M | #28 of 89 | Archive leaderboard | report |
| Language Modelling | WikiText-103 | Sandwich Transformer | Test perplexity | 17.96 | #28 of 89 | Archive leaderboard | report |
| Language Modelling | enwik8 | Sandwich Transformer (adaptive span) | Bit per Character (BPC) | 0.968 | #7 of 42 | Archive leaderboard | report |
| Language Modelling | enwik8 | Sandwich Transformer (adaptive span) | Number of params | 209M | #7 of 42 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Introduced by this paper: Sandwich Transformer
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections