SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.00175 · 2023

Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblocks

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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, 1 violated it and 0 have no contract verdict; 7 failures of our instrument; 0 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/2307.00175.json · badge: /api/ran/2307.00175.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
balevinstein/probes canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
5 ran · 0 no recorded run
FunctionStatusWhere it lives
get_embeddings Ran
ran; our contract draft was wrong, not the code
balevinstein/probes/Train_CCSProbe.py
code served (permissive licence) · get_code("1fd11bf75e70a151")
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 tuple. 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 torch
import pandas as pd
import numpy as np

torch.manual_seed(0)

# Create a small DataFrame with the structure expected by get_embeddings
# The function expects data.iloc[idx] to return rows with 'embeddings' and 'label' columns
data_list = []
for i in range(4):  # Even number of rows as required by assertion
    data_list.append({
        'embeddings': torch.randn(4),  # Small embedding dimension
        'label': i % 2
    })

data = pd.DataFrame(data_list)

ARGS = {
    'data': data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
load_config Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from balevinstein/probes
pointer only · get_code("47a75377f206e2cb")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

load_csv Ran
ran; our contract draft was wrong, not the code
balevinstein/probes/Train_CCSProbe.py
code served (permissive licence) · get_code("490da4aec73c34bb")
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 DataFrame. 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 pandas as pd
import tempfile
import os

# Create a temporary CSV file with embeddings data
with tempfile.NamedTemporaryFile(mode='w', suffix='.csv', delete=False) as f:
    temp_filename = f.name
    f.write("id,embeddings,label\n")
    f.write("1,\"[0.1, 0.2, 0.3, 0.4]\",A\n")
    f.write("2,\"[0.5, 0.6, 0.7, 0.8]\",B\n")

ARGS = {
    'filename': temp_filename
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
load_data Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from balevinstein/probes
pointer only · get_code("f2ad4d1f97c24685")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

load_datasets Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from balevinstein/probes
pointer only · get_code("c32cb2b828044b4d")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

prepare_data_for_mlp Ran
ran, violated the contract we drafted
balevinstein/probes/Train_CCSProbe.py
code served (permissive licence) · get_code("f2d39d520a929049")
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 tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 torch
import numpy as np

torch.manual_seed(0)

# Create small sample embeddings for positive examples
# Shape: (batch_size=2, embedding_dim=4)
all_pos_hs = [
    [0.1, 0.2, 0.3, 0.4],
    [0.5, 0.6, 0.7, 0.8]
]

# Create small sample embeddings for negative examples
# Shape: (batch_size=2, embedding_dim=4)
all_neg_hs = [
    [0.9, 0.8, 0.7, 0.6],
    [0.5, 0.4, 0.3, 0.2]
]

ARGS = {
    "all_pos_hs": all_pos_hs,
    "all_neg_hs": all_neg_hs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
prepare_datasets Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from balevinstein/probes
pointer only · get_code("5b5f6a3969b72cda")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

process_row Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from balevinstein/probes
pointer only · get_code("cc6f60b3b538e888")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 8 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 8 later) and 0 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
load_config 2 other papers
1 dated earlier, 1 later
2304.13734 (2023-04)
The Internal State of an LLM Knows When It's Lying
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
load_data 2 other papers
1 dated earlier, 1 later
2304.13734 (2023-04)
The Internal State of an LLM Knows When It's Lying
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
load_datasets 2 other papers
1 dated earlier, 1 later
2304.13734 (2023-04)
The Internal State of an LLM Knows When It's Lying
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
prepare_datasets 2 other papers
1 dated earlier, 1 later
2304.13734 (2023-04)
The Internal State of an LLM Knows When It's Lying
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
process_row 2 other papers
1 dated earlier, 1 later
2304.13734 (2023-04)
The Internal State of an LLM Knows When It's Lying
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
get_embeddings 1 other paper
1 later
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
load_csv 1 other paper
1 later
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics
prepare_data_for_mlp 1 other paper
1 later
2024.findings-naacl.294 (2024 (year only))
PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics

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

We consider the questions of whether or not large language models (LLMs) have beliefs, and, if they do, how we might measure them. First, we evaluate two existing approaches, one due to Azaria and Mitchell (2023) and the other to Burns et al. (2022). We provide empirical results that show that these methods fail to generalize in very basic ways. We then argue that, even if LLMs have beliefs, these methods are unlikely to be successful for conceptual reasons. Thus, there is still no lie-detector for LLMs. After describing our empirical results we take a step back and consider whether or not we should expect LLMs to have something like beliefs in the first place. We consider some recent arguments aiming to show that LLMs cannot have beliefs. We show that these arguments are misguided. We provide a more productive framing of questions surrounding the status of beliefs in LLMs, and highlight the empirical nature of the problem. We conclude by suggesting some concrete paths for future work.

For agents

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

get_harvested_code_for_paper("2307.00175")
get_code_for_paper("2307.00175")
have("2307.00175")

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

curl https://syntology.ai/api/ran/2307.00175.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.00175.svg)](https://syntology.ai/paper/2307.00175)

Connect an agent — have() is free.