Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-06T21:23:40.653122Z
Paper Citation Record · LEDGER
As of 8 August 2026, this Paper Citation Record lists 47 of 47 outbound references and 0 inbound Pith citation observations for arXiv:2507.00322.
A citation records a reference. It does not transfer a finding from one paper to another.
Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-06T21:23:40.653122Z
One-hop event checks from named stored sources.
Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+00:00
Pith citing papers itemized under the disclosed page cap.
Source: paper_references, paper_reference_links
A source-named dated measurement, never combined with another source.
Source: cited_works
47 of 47 outbound references displayed
External citation measurements
No source-named external measurement is stored.
Observation 5e856afd-33f1-4e5a-9952-bfe41eddfad3 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
Reference 1
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation a535a966-6130-4fbd-a6fa-21b2d11f48d6 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones GPT-4 Technical Report
Reference 2
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 0b62b822-411b-4e50-be6c-d0ecff9fa0b1 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Refusal in Language Models Is Mediated by a Single Direction
Reference 3
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation c5912ac0-7a11-41db-ab5f-db247c0a94bf · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Program Synthesis with Large Language Models
Reference 4
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f54fe905-c6e1-4591-b2a6-6a4e6e738f2f · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Mechanistic interpretability for AI safety - a re- view.Transactions on Machine Learning Research, 2024
Reference 5
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation c5aedd5b-535c-4bf5-85a6-aaa96562c651 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Pythia: A suite for analyzing large language models across training and scaling
Reference 6
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 6a5d405e-a3d7-4997-b08b-8b9496445206 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Towards monosemanticity: Decomposing language models with dictionary learning.Transformer Circuits Thread, 2023
Reference 7
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation e141caff-ef8c-4178-859b-c7725809b015 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Causal scrubbing: A method for rigorously testing interpretability hypotheses
Reference 8
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation ff537701-90ea-4075-abfe-eb5453d7172e · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Unresolved cited work
Reference 9
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 8e651267-9ff8-4c1e-bf5e-47a273ed6c41 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Summing Up the Facts: Additive Mechanisms Behind Factual Recall in LLMs
Reference 10
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4bea593a-9fa0-43a0-b6e7-132a52660b25 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Analyzing Transformers in Embedding Space
Reference 11
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 2770c8af-cfe3-4ad7-8a32-eeb6bef9b0c7 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones What’s wrong with your code generated by large language models? an extensive study.arXiv preprint arXiv:2407.06153, 2024
Reference 12
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 5f928bd4-3bd9-467a-8a2c-34616035bfe7 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones The llama 3 herd of models.arXiv e-prints, pages arXiv–2407, 2024
Reference 13
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation deca5664-4f0a-45e6-8967-c3901ecd20d5 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones A mathematical framework for transformer circuits.Transformer Circuits Thread,
Reference 14
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 4d6e15ee-1f1a-4c5f-8aaf-35c71ffeaa7b · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones InCoder: A Generative Model for Code Infilling and Synthesis
Reference 15
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 5831349c-d295-444f-b1b7-462875a4b994 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
Reference 16
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 0922e5da-b6bf-496f-a2cc-c1a22bcf0f98 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones The Llama 3 Herd of Models
Reference 17
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 86c5e29a-1c40-41b9-80f4-611181896140 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones How does gpt-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained language model.Advances in Neural Information Processing Systems, 36:76033–76060, 2023
Reference 18
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 189f0de1-afd7-4d09-83df-eadfb4e688d1 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Have faith in faithfulness: Going be- yond circuit overlap when finding model mechanisms
Reference 19
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 73480e50-707f-49ea-9b18-8749d2e99412 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones How to use and interpret activation patching
Reference 20
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f5664269-a8d1-4a1a-a51a-2b8296c8987e · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Livecodebench: Holistic and contamination free evaluation of large language models for code.CoRR, 2024
Reference 21
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 25b3e83d-6176-4e35-940d-c901c6fd6c1f · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Inference- time intervention: Eliciting truthful answers from a language model.Advances in Neural Information Processing Systems, 36:41451–41530, 2023
Reference 22
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation c932472a-f480-411a-9ee6-00cd753ee253 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Starcoder: may the source be with you!Transactions on Machine Learning Research, 2023
Reference 23
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 91125b34-96a4-48c2-bd86-e36bdb4d8e73 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Unresolved cited work
Reference 24
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4b2c5672-c6e4-43e1-946b-0f72f9ca9025 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones In-context vectors: making in context learning more effective and controllable through latent space steering
Reference 25
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 6a4ab492-ec66-45b5-bd28-08b12e1b8253 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Mechanistic Understanding of Language Models in Syntactic Code Completion
Reference 26
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation a4c3e5f5-a8c7-4a92-91d8-7bae60c8462d · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones CodeGen2: Lessons for Training LLMs on Programming and Natural Languages
Reference 27
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation e01e7cb0-6b59-4ba9-a811-f362af5b0bf8 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Arithmetic without algo- rithms: Language models solve math with a bag of heuristics
Reference 28
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation a181ac3f-1dcf-4626-87c0-045e527f6b99 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Interpreting gpt: the logit lens.AI Alignment Forum, 2020
Reference 29
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 1688a73a-d51a-448f-a014-39ddc0e72272 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Zoom in: An introduction to circuits.Distill, 5(3):e00024–001, 2020
Reference 30
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation ac528d42-7175-4088-8bbc-08b87c788ba5 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Language models are unsupervised multitask learners.OpenAI blog, 1(8):9, 2019
Reference 31
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4b2be69a-3714-4a37-9890-1a3f9ce5afb0 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones An investigation of neuron activation as a unified lens to explain chain-of-thought eliciting arithmetic reasoning of llms
Reference 32
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 75dfdb9f-ce66-4464-9d8e-1fb16f1ed617 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones A practical review of mech- anistic interpretability for transformer-based language models.arXiv preprint arXiv:2407.02646, 2024
Reference 33
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 82549726-e047-4666-a98a-2264d29c3437 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Red-teaming language models via activation engineering.Accessed: October, 13:2023, 2023
Reference 34
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 9891157f-3740-44ab-8e5d-c1ed42a41cbf · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Steering llama 2 via contrastive activation addition
Reference 35
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 3edd397d-5696-40d1-b388-fd01b7e894df · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Code Llama: Open Foundation Models for Code
Reference 36
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 533344cc-2aa2-4f27-a1ef-38a631d0c7d7 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Attribution patching outperforms automated circuit discovery
Reference 37
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f6357461-1fef-439b-b87d-65af4519cad7 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Linear representations of sentiment in large language models.CoRR, 2023
Reference 38
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 2ad534bc-78ec-49c3-8e0b-fd5dec4098ff · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Llama 2: Open Foundation and Fine-Tuned Chat Models
Reference 39
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 8359c874-8dc3-4271-88a4-15e11a3f0309 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Activation addition: Steering language models without optimization.CoRR, 2023
Reference 40
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 34339abf-a0ed-49e0-9a83-ceb3bba8cca8 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Extending Activation Steering to Broad Skills and Multiple Behaviours
Reference 41
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 3625b138-83b3-4f16-8f78-371a3a288972 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Attention is all you need.Advances in neural information processing systems, 30, 2017
Reference 42
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f95f3fc4-d93b-41ac-835f-39db7d4b9be8 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Interpretability in the wild: a circuit for indirect object identification in GPT-2 small
Reference 43
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation a70ee580-114b-4c14-b238-1cb1f7aa4d2d · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Towards Understanding the Characteristics of Code Generation Errors Made by Large Language Models
Reference 44
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation e493e9e3-6a26-48e1-9a6b-e5092049b00b · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Reft: Representation finetuning for language models.Advances in Neural Information Processing Systems, 37:63908–63962, 2024
Reference 45
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 179d2a8f-c581-409d-9322-eafce77aa6f1 · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones #print the string 160\nprint(str(160
Reference 46
Source-reported events for the cited work
No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.
Observation 7b7c97ba-0bd8-4f50-8f3b-0eb864cdc40d · outbound
Failure by Interference: Language Models Make Balanced Parentheses Errors When Faulty Mechanisms Overshadow Sound Ones Unresolved cited work
Reference 2021
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
No inbound Pith citation observations are available.