Papers › Revisiting In-context Learning Inference Circuit in Large Language Models

Revisiting In-context Learning Inference Circuit in Large Language Models

6 Oct 2024arXiv:2410.04468archive 2025-07-28

Hakaze Cho, Mariko Kato, Yoshihiro Sakai, Naoya Inoue

In-context Learning (ICL) is an emerging few-shot learning paradigm on Language Models (LMs) with inner mechanisms un-explored. There are already existing works describing the inner processing of ICL, while they struggle to capture all the inference phenomena in large language models. Therefore, this paper proposes a comprehensive circuit to model the inference dynamics and try to explain the observed phenomena of ICL. In detail, we divide ICL inference into 3 major operations: (1) Input Text Encode: LMs encode every input text (demonstrations and queries) into linear representation in the hidden states with sufficient information to solve ICL tasks. (2) Semantics Merge: LMs merge the encoded representations of demonstrations with their corresponding label tokens to produce joint representations of labels and demonstrations. (3) Feature Retrieval and Copy: LMs search the joint representations similar to the query representation on a task subspace, and copy the searched representations into the query. Then, language model heads capture these copied label representations to a certain extent and decode them into predicted labels. The proposed inference circuit successfully captured many phenomena observed during the ICL process, making it a comprehensive and practical explanation of the ICL inference process. Moreover, ablation analysis by disabling the proposed steps seriously damages the ICL performance, suggesting the proposed inference circuit is a dominating mechanism. Additionally, we confirm and list some bypass mechanisms that solve ICL tasks in parallel with the proposed circuit.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 15 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 15 samples from 1 repository, 11 ran (11 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

hc495/ICL_Circuit officialmentioned on GitHubpytorchMITSyntology: 15 samples, 11 ran (0 instrument-failed) · 4 unverified report

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

15 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
10ran
4unverified

Licence: 0 of the 15 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from hc495/ICL_Circuit. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

hc495/ICL_Circuit official repository 15 samples, 11 ran (0 instrument-failed) · 4 unverified
ICL_inference_to_hidden_states util/inference.py ran MIT (permissive) · 6d2ba9201e0266c1 · report fixture: signature model:callable/0/?, tokenizer:callable/0/?, prompts:list/1/str · returned list · level recorded 2026-09-24
ICL_inference_to_hidden_states_transposed util/inference.py ran MIT (permissive) · 0e4a9d80be097eab · report fixture: signature model:callable/0/?, tokenizer:callable/0/?, prompts:list/1/str · returned list · level recorded 2026-09-24
encoder_inference_to_feature util/inference.py ran MIT (permissive) · 03f5ae8eefe2ed8f · report fixture: signature model:callable/0/?, tokenizer:callable/0/?, queries:list/1/str · returned list · level recorded 2026-09-24
find_tokenized_label_word util/ablation_study.py ran MIT (permissive) · dc2a498623d25b35 · report fixture: signature tokenizer:callable/0/?, experimentor:MockExperimentor/0/?, prompt:str/0/str · returned tuple · level recorded 2026-09-24
get_induction_correctness_for_single_layer util/induction_evaluation.py ran MIT (permissive) · ab0facc671844bd0 · report fixture: signature ICL_attention:dict/0/?, experimentor:MockExperimentor/0/?, sample_index:int/0/int, layer:int/0/int · returned list · level recorded 2026-09-24
get_induction_magnitude_for_single_layer util/induction_evaluation.py ran MIT (permissive) · 36830c7e7fd81df1 · report fixture: signature ICL_attention:list/1/list, experimentor:Experimentor/0/?, sample_index:int/0/int, layer:int/0/int · returned list · level recorded 2026-09-24
kernel_alignment util/kernel_alignment.py ran MIT (permissive) · 61a2e6dfc5a9fe1f · report fixture: signature simGraph_1:ndarray/2/float, simGraph_2:ndarray/2/float, k:int/0/int · returned tuple · level recorded 2026-09-24
load_data_from_StaICC_experimentor util/load_model_and_data.py ran MIT (permissive) · 638cfdf35bdeb6ba · report fixture: signature experimentor:MockExperimentor/0/?, prompt_cut:str/0/str, target_label_correction:bool/0/bool · returned tuple · level recorded 2026-09-24
overlap in 7 papers util/kernel_alignment.py ran fingerprinted MIT (permissive) · 5d93a88b24afc491 · report fixture: signature a:list/1/int, b:list/1/int · returned int · level recorded 2026-09-24
sim_graph util/kernel_alignment.py ran · honoured contract MIT (permissive) · 81967b2af6f55803 · report fixture: signature features:ndarray/2/float · returned list · level recorded 2026-09-24
tokenized_length util/induction_evaluation.py ran MIT (permissive) · 3a45ae8b2ab1d6e9 · report fixture: signature tokenizer:callable/0/?, prompt:str/0/str · returned int · level recorded 2026-09-24
Masked_ICL_inference util/ablation_study.py unverified MIT (permissive) · 8ac576595ccbc620 · report
head_ablation util/ablation_study.py unverified MIT (permissive) · ab89dd08b607b43c · report
load_ICL_model util/load_model_and_data.py unverified MIT (permissive) · f906d6d0f7a495ad · report
load_encode_model util/load_model_and_data.py unverified MIT (permissive) · fdf0bd39d2cf5fed · report

Tasks

Few-Shot LearningIn-Context Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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