Paper Citation Record · LEDGER

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model

As of 19 August 2026, this Paper Citation Record lists 35 of 35 outbound references and 3 inbound Pith citation observations for arXiv:2504.13292.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2504.13292 v1

Coverage vector

measured 35 of 35 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T12:18:39.322958Z

measured 38 of 38 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-19T06:32:44.657259+00:00

measured 3 of 3 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T15:20:33.165073Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: pith, observed 2026-07-11T01:17:45.447627Z

Reference resolution

35 of 35 outbound references displayed

verified exact0
verified fuzzy13
unresolved22
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 49a4fe7f-31bc-4ec0-92c4-69e087c42227 · outbound

This paper cites We condition on the event ∥z− ¯z∥≤ ε2 rp n logn.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model We condition on the event ∥z− ¯z∥≤ ε2 rp n logn

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:40.000319Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 4942872c-159c-44db-b9ab-94188db87502 · outbound

This paper cites Denote the signal ofzi by ¯zi = [µ2,−µ2,−µ1]⊤¯xi.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Denote the signal ofzi by ¯zi = [µ2,−µ2,−µ1]⊤¯xi

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:40.056338Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation f16e7b0d-fee2-4321-9053-696c0da38b4c · outbound

This paper cites Grokking Modular Polynomials.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokking Modular Polynomials

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.531824Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a20c3aac-93e0-4968-9911-9522a5b7ebc0 · outbound

This paper cites Learning to grok: Emergence of in-context learning and skill composition in modular arithmetic tasks.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Learning to grok: Emergence of in-context learning and skill composition in modular arithmetic tasks

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.548622Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 54c68532-1883-4e47-ac61-4564cddd9093 · outbound

This paper cites Deep Networks Always Grok and Here is Why.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Deep Networks Always Grok and Here is Why

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.554591Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05e8e003-7082-49b4-9fa7-b68b4c59ef90 · outbound

This paper cites Scaling Laws for Neural Language Models.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Scaling Laws for Neural Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.559605Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0bc2c8ae-f5fd-43bc-9a27-c036ee3aaeee · outbound

This paper cites Grokfast: Accelerated Grokking by Amplifying Slow Gradients.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokfast: Accelerated Grokking by Amplifying Slow Gradients

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.564442Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 988c9ca1-5175-4396-867f-e8ac45c04d5f · outbound

This paper cites Decoupled weight decay regularization.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Decoupled weight decay regularization

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:40.274343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 4cbbb038-2b2c-4275-bac6-0666883a1e69 · outbound

This paper cites Emergence in non-neural models: grokking modular arithmetic via average gradient outer product.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Emergence in non-neural models: grokking modular arithmetic via average gradient outer product

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.573985Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f04ee7fc-58c7-4cd4-bdaa-8e1fbfc2c282 · outbound

This paper cites A Tale of Two Circuits: Grokking as Competition of Sparse and Dense Subnetworks.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model A Tale of Two Circuits: Grokking as Competition of Sparse and Dense Subnetworks

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.579151Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0584ffa4-507f-4707-b536-b563278075d1 · outbound

This paper cites Grokking Beyond Neural Networks: An Empirical Exploration with Model Complexity.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokking Beyond Neural Networks: An Empirical Exploration with Model Complexity

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.584329Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ba53f8ba-1d95-434a-b7fe-5f25f701077d · outbound

This paper cites an unresolved cited work.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-08-16T12:18:40.259375Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 762cf8ac-993b-42d7-95d0-514689977305 · outbound

This paper cites Why Do You Grok? A Theoretical Analysis of Grokking Modular Addition.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Why Do You Grok? A Theoretical Analysis of Grokking Modular Addition

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.593849Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 180193fc-e8bf-4a8a-807c-31bb59d0b8ca · outbound

This paper cites Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.636026Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d9681e2-4fd3-49cb-9476-5d98426dc4fb · outbound

This paper cites Explaining grokking through circuit efficiency.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Explaining grokking through circuit efficiency

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.884568Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 96dabb3c-ab02-4be5-847f-7a68d6c7ee16 · outbound

This paper cites Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.889675Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 63389e8c-c0e2-4cec-843d-76464517a6c8 · outbound

This paper cites Learning to grow pretrained models for efficient transformer training.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Learning to grow pretrained models for efficient transformer training

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:40.245664Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 57b08f18-38e9-4d49-a06d-0b6eeb4a3015 · outbound

This paper cites Critical Data Size of Language Models from a Grokking Perspective.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Critical Data Size of Language Models from a Grokking Perspective

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.899644Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 37776244-81ab-4388-9295-99fc11cd386d · outbound

This paper cites Grokking phase transitions in learning local rules with gradient descent.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokking phase transitions in learning local rules with gradient descent

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.903816Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a239f32d-aa77-4367-85a7-d4640ca6af36 · outbound

This paper cites 15 A.1.1 Proof of Lemma 3.1.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model 15 A.1.1 Proof of Lemma 3.1

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:40.134929Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation c20f191b-3ca7-4c45-adb8-fa24eccd3ab5 · outbound

This paper cites Then ⟨v(1) j ,z⟩ < 0 also hold for aj < 0 following the same analysis.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Then ⟨v(1) j ,z⟩ < 0 also hold for aj < 0 following the same analysis

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:40.016885Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation bbcbc072-00e5-4bc2-9274-941799b6caf9 · outbound

This paper cites an unresolved cited work.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Unresolved cited work

Reference 27

Resolution: unresolved; raw_fallback, observed 2026-08-16T12:18:39.984851Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 45d281c4-b5c3-4740-a3b0-537ca5a5f4e7 · outbound

This paper cites Similarly we have P( |I[1,1,1],−µ2|− |I−µ2| 2 >t )≤ 2 exp(−2t2 n ).

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Similarly we have P( |I[1,1,1],−µ2|− |I−µ2| 2 >t )≤ 2 exp(−2t2 n )

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.970090Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 6e9ea831-d41f-4b09-b558-54c9686538d9 · outbound

This paper cites Figure 5a generates 4000 i.i.d datapoints from the distribution P , and visualizes Ux for each x.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Figure 5a generates 4000 i.i.d datapoints from the distribution P , and visualizes Ux for each x

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.939155Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation e377f38b-d74c-4f95-84b1-f4b0e5e47435 · outbound

This paper cites an unresolved cited work.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Unresolved cited work

Reference 31

Resolution: unresolved; raw_fallback, observed 2026-08-16T12:18:39.735193Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation a371782a-c1b4-4b59-bfa4-572a21393f30 · outbound

This paper cites The optimal configuration for GrokFast is (lr, wd) = (0.01, 1.0).

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model The optimal configuration for GrokFast is (lr, wd) = (0.01, 1.0)

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.682731Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 69cd792e-fb0c-4c0b-966d-3cc878136661 · outbound

This paper cites (2020), we have N = 2dembednlayer(2dattn+dmlp) = 2∗512∗8∗(256+512) = 6291456,Cforward = 2(N+8∗2∗128)∼.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model (2020), we have N = 2dembednlayer(2dattn+dmlp) = 2∗512∗8∗(256+512) = 6291456,Cforward = 2(N+8∗2∗128)∼

Reference 103

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.667391Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 7b84ac13-2b30-4d4c-ad20-2d4343c995a1 · outbound

This paper cites Target model trained by GrokTransfer takes around 1000 epochs and target model trained from scratch takes around 10000 epochs.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Target model trained by GrokTransfer takes around 1000 epochs and target model trained from scratch takes around 10000 epochs

Reference 107

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.653014Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation d9a62d74-5f48-4ef8-835d-a750a2e6da7c · outbound

This paper cites We select the configuration that first achieves 90% accuracy on the validation set.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model We select the configuration that first achieves 90% accuracy on the validation set

Reference 512

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.954880Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation c677da32-9b64-4fbc-9c76-4068176286df · outbound

This paper cites Middle: Visualization of the neuron in the weak model.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Middle: Visualization of the neuron in the weak model

Reference 1011

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:18:39.638086Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation 0dc2ff5e-bec7-4b33-812f-2e61b89aa064 · outbound

This paper cites The Slingshot Mechanism: An Empirical Study of Adaptive Optimizers and the Grokking Phenomenon.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model The Slingshot Mechanism: An Empirical Study of Adaptive Optimizers and the Grokking Phenomenon

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.775846Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d60214f-b51f-44cb-87d2-8a63852ad5c5 · outbound

This paper cites Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.347645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dbbd0480-111d-4f77-940a-bc93b5f14880 · outbound

This paper cites Unifying Grokking and Double Descent.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Unifying Grokking and Double Descent

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.438252Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a77395fb-5519-4796-af4c-bceb0081d27c · outbound

This paper cites Towards Empirical Interpretation of Internal Circuits and Properties in Grokked Transformers on Modular Polynomials.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Towards Empirical Interpretation of Internal Circuits and Properties in Grokked Transformers on Modular Polynomials

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.537618Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 75d87e79-5422-4e5f-ad52-5a58adc69199 · outbound

This paper cites Grokking modular arithmetic.

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model Grokking modular arithmetic

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-16T12:18:38.543229Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation befd7f07-cf2f-4dd4-837c-c158b116bf73 · inbound

Mechanistic Insights into Grokking from the Embedding Layer cites this paper.

Mechanistic Insights into Grokking from the Embedding Layer Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-07T15:20:33.165073Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7dec3de9-3a91-4c6a-9c95-4eade5a0dfc5 · inbound

Cross-Trajectory Chimera Interventions Reveal Dissociable Roles of Weight Magnitude and Direction in Grokking cites this paper.

Cross-Trajectory Chimera Interventions Reveal Dissociable Roles of Weight Magnitude and Direction in Grokking Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model

Reference 5

Resolution: metadata mismatch; local_arxiv, observed 2026-07-11T01:17:45.473343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.

Observation c6f7e72b-74c0-4d14-8dc1-0b2ba8d0cdcb · inbound

How to Tame Grokking: Representation Geometry as a Control Signal cites this paper.

How to Tame Grokking: Representation Geometry as a Control Signal Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model

Reference 28

Resolution: malformed identifier; no resolver link, observed 2026-07-14T04:01:12.361370Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.