SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.09635 · ICML · 2024

Transformers Get Stable: An End-to-End Signal Propagation Theory for Language Models

Akhil Kedia, Abbas Mohd, Zaidi, Sushil Khyalia, Jungho Jung, Harshith Goka, Haejun Lee

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2403.09635.json · badge: /api/ran/2403.09635.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
akhilkedia/tranformersgetstable canonical (classifier label)
classifier basis: census
4 ran · 1 no recorded run
FunctionStatusWhere it lives
calc_err Ran
ran, honoured the contract we drafted
akhilkedia/tranformersgetstable/simulations/layernorm.py
pointer only (licence: NONE) · get_code("439e1fb57ee36d78")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, akhilkedia/tranformersgetstable when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_batch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
akhilkedia/tranformersgetstable/DSLM_signal_propagation_figures/pretrain_vision_classify.py
pointer only (licence: NONE); the stored copy get_code serves, from Centaurus-Alpha/UniPool, is pointer only (licence: NOASSERTION) · get_code("1f4447581128a711")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, akhilkedia/tranformersgetstable when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_batch Ran
ran; our contract draft was wrong, not the code
akhilkedia/tranformersgetstable/DSLM_signal_propagation_figures/pretrain_vision_dino.py
pointer only (licence: NONE); the stored copy get_code serves, from Centaurus-Alpha/UniPool, is pointer only (licence: NOASSERTION) · get_code("92287bdc91b8ac90")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, akhilkedia/tranformersgetstable when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_batch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
akhilkedia/tranformersgetstable/DSLM_signal_propagation_figures/pretrain_vision_inpaint.py
pointer only (licence: NONE); the stored copy get_code serves, from Centaurus-Alpha/UniPool, is pointer only (licence: NOASSERTION) · get_code("edf213e729a014e0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, akhilkedia/tranformersgetstable when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

simulate_one Not yet run
unverified (harvested, no recorded run)
akhilkedia/tranformersgetstable/simulations/layernorm.py
pointer only (licence: NONE) · get_code("44194e1b32ba91a3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, akhilkedia/tranformersgetstable when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 3 the same month, 3 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_batch 8 other papers
1 dated earlier, 1 the same month, 6 later
2312.04916 (2023-12)
EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism
2403.18926 (2024-03)
XMoE: Sparse Models with Fine-grained and Adaptive Expert Selection
2405.17976 (2024-05)
Yuan 2.0-M32: Mixture of Experts with Attention Router
and 5 more not named here
get_batch 8 other papers
1 dated earlier, 1 the same month, 6 later
2312.04916 (2023-12)
EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism
2403.18926 (2024-03)
XMoE: Sparse Models with Fine-grained and Adaptive Expert Selection
2405.17976 (2024-05)
Yuan 2.0-M32: Mixture of Experts with Attention Router
and 5 more not named here
get_batch 7 other papers
1 the same month, 6 later
2403.18926 (2024-03)
XMoE: Sparse Models with Fine-grained and Adaptive Expert Selection
2405.17976 (2024-05)
Yuan 2.0-M32: Mixture of Experts with Attention Router
2411.03884 (2024-11)
Polynomial Composition Activations: Unleashing the Dynamics of Large Language Models
and 4 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In spite of their huge success, transformer models remain difficult to scale in depth. In this work, we develop a unified signal propagation theory and provide formulae that govern the moments of the forward and backward signal through the transformer model. Our framework can be used to understand and mitigate vanishing/exploding gradients, rank collapse, and instability associated with high attention scores. We also propose Deep-ScaleLM, an initialization and scaling scheme that conserves unit output/gradient moments throughout the model, enabling the training of very deep models with 1000 layers. We find that transformer models could be much deeper -our deep models with fewer parameters outperform shallow models in Language Modeling, Speech Translation, and Image Classification, across encoder-only, decoder-only and encoder-decoder variants, for both Pre-LN and Post-LN transformers, for multiple datasets and model sizes. These improvements also translate into improved performance on downstream Question Answering tasks and improved robustness for Image Classification.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2403.09635")
get_code_for_paper("2403.09635")
have("2403.09635")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2403.09635.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2403.09635.svg)](https://syntology.ai/paper/2403.09635)

Connect an agent — have() is free.