SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.00799 · 2024

Get my drift? Catching LLM Task Drift with Activation Deltas

arXiv · PDF · Open in the Atlas

Code that ran

1 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 4 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/2406.00799.json · badge: /api/ran/2406.00799.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
microsoft/TaskTracker canonical (classifier label)
classifier basis: census
1 ran · 4 no recorded run
FunctionStatusWhere it lives
load_config Ran
ran; our contract draft was wrong, not the code
microsoft/TaskTracker/quick_start/utils.py
code served (permissive licence) · get_code("24419cde3a1d3292")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import logging
import tempfile
import os
import yaml

# Set up logging to suppress output during test
logging.basicConfig(level=logging.CRITICAL)

# Create a temporary YAML config file
temp_dir = tempfile.mkdtemp()
config_file = os.path.join(temp_dir, 'config.yaml')

# Write a valid YAML config
config_data = {
    'model': 'test_model',
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 10,
    'layers': [64, 128, 256]
}

with open(config_file, 'w') as f:
    yaml.dump(config_data, f)

ARGS = {
    'config_path': config_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_last_token_activations_single Not yet run
unverified (harvested, no recorded run)
microsoft/TaskTracker/task_tracker/utils/activations.py
code served (permissive licence) · get_code("42886e30b586b4c9")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
microsoft/TaskTracker/task_tracker/utils/model.py
code served (permissive licence) · get_code("1d64b715a817507d")
How it ran

no invocation recorded.

load_task_tracker Not yet run
unverified (harvested, no recorded run)
microsoft/TaskTracker/quick_start/utils.py
code served (permissive licence) · get_code("095aafd28c666431")
How it ran

no invocation recorded.

setup_hf_llm Not yet run
unverified (harvested, no recorded run)
microsoft/TaskTracker/quick_start/utils.py
code served (permissive licence) · get_code("93318919061761ca")
How it ran

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, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 5 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.

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

LLMs are commonly used in retrieval-augmented applications to execute user instructions based on data from external sources. For example, modern search engines use LLMs to answer queries based on relevant search results; email plugins summarize emails by processing their content through an LLM. However, the potentially untrusted provenance of these data sources can lead to prompt injection attacks, where the LLM is manipulated by natural language instructions embedded in the external data, causing it to deviate from the user's original instruction(s). We define this deviation as task drift. Task drift is a significant concern as it allows attackers to exfiltrate data or influence the LLM's output for other users. We study LLM activations as a solution to detect task drift, showing that activation deltas - the difference in activations before and after processing external data - are strongly correlated with this phenomenon. Through two probing methods, we demonstrate that a simple linear classifier can detect drift with near-perfect ROC AUC on an out-of-distribution test set. We evaluate these methods by making minimal assumptions about how users' tasks, system prompts, and attacks can be phrased. We observe that this approach generalizes surprisingly well to unseen task domains, such as prompt injections, jailbreaks, and malicious instructions, without being trained on any of these attacks. Interestingly, the fact that this solution does not require any modifications to the LLM (e.g., fine-tuning), as well as its compatibility with existing meta-prompting solutions, makes it cost-efficient and easy to deploy. To encourage further research on activation-based task inspection, decoding, and interpretability, we release our large-scale TaskTracker toolkit, featuring a dataset of over 500K instances, representations from six SoTA language models, and a suite of inspection tools.

For agents

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

get_harvested_code_for_paper("2406.00799")
get_code_for_paper("2406.00799")
have("2406.00799")

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

curl https://syntology.ai/api/ran/2406.00799.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.00799.svg)](https://syntology.ai/paper/2406.00799)

Connect an agent — have() is free.