SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2409.17027 · 2024

Counterfactual Token Generation in Large Language Models

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 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; 2 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/2409.17027.json · badge: /api/ran/2409.17027.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
networks-learning/counterfactual-llms canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
apply_rotary_emb Ran
ran; our contract draft was wrong, not the code
networks-learning/counterfactual-llms/src/mistral-inference/moe_one_file_ref.py
pointer only (licence: NONE) · get_code("22426d7eecb70410")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

fix_income Ran
ran, honoured the contract we drafted
networks-learning/counterfactual-llms/src/bias.py
pointer only (licence: NONE) · get_code("7e3a2dbb3c9579a1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

get_fig_dim Ran
ran, honoured the contract we drafted
networks-learning/counterfactual-llms/src/utils.py
pointer only (licence: NONE) · get_code("1c1a16c4896a4f25")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

get_where_to_intervene_dict Ran
ran; our contract draft was wrong, not the code
networks-learning/counterfactual-llms/src/bias.py
pointer only (licence: NONE) · get_code("7c2a58be8462ca91")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

get_where_to_intervene_dict_direct Ran
ran; our contract draft was wrong, not the code
networks-learning/counterfactual-llms/src/bias.py
pointer only (licence: NONE) · get_code("58d741b58f7d0059")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

precompute_freqs_cis Ran
ran, violated the contract we drafted
networks-learning/counterfactual-llms/src/mistral-inference/moe_one_file_ref.py
pointer only (licence: NONE) · get_code("99e0ca21ba709107")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

repeat_kv Ran
ran; our contract draft was wrong, not the code
networks-learning/counterfactual-llms/src/mistral-inference/moe_one_file_ref.py
pointer only (licence: NONE) · get_code("3fdd4f66367ca608")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

repeat_kv Ran
ran; our contract draft was wrong, not the code
networks-learning/counterfactual-llms/src/mistral-inference/one_file_ref.py
pointer only (licence: NONE) · get_code("e8e39c9933cb102c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

reshape_for_broadcast Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
networks-learning/counterfactual-llms/src/llama3/llama/model.py
pointer only (licence: NONE); the stored copy get_code serves, from alpha-vllm/llama2-accessory, is pointer only (licence: NOASSERTION) · get_code("70bf6ebaafd266c4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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.

sample_top_p Ran
ran; our contract draft was wrong, not the code
networks-learning/counterfactual-llms/src/llama3/llama/generation.py
pointer only (licence: NONE) · get_code("841e0eb44df294a1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, networks-learning/counterfactual-llms when it was checked. 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 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 8 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
reshape_for_broadcast 58 other papers
38 dated earlier, 20 later · 2 only as an assembly brick
2023.emnlp-main.534 (2023 (year only))
Disentangling Transformer Language Models as Superposed Topic Models
2303.09736 (2023-03)
Dynamic Structure Pruning for Compressing CNNs
2307.09288 (2023-07)
Llama 2: Open Foundation and Fine-Tuned Chat Models
and 55 more not named here
precompute_freqs_cis 1 other paper
1 dated earlier
2310.06825 (2023-10)
Mistral 7B

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

"Sure, I am happy to generate a story for you: Captain Lyra stood at the helm of her trusty ship, the Maelstrom's Fury, gazing out at the endless sea. [...] Lyra's eyes welled up with tears as she realized the bitter truth - she had sacrificed everything for fleeting riches, and lost the love of her crew, her family, and herself." Although this story, generated by a large language model, is captivating, one may wonder -- how would the story have unfolded if the model had chosen "Captain Maeve" as the protagonist instead? We cannot know. State-of-the-art large language models are stateless -- they maintain no internal memory or state. Given a prompt, they generate a sequence of tokens as an output using an autoregressive process. As a consequence, they cannot reason about counterfactual alternatives to tokens they have generated in the past. In this work, our goal is to enhance them with this functionality. To this end, we develop a causal model of token generation that builds upon the Gumbel-Max structural causal model. Our model allows any large language model to perform counterfactual token generation at almost no cost in comparison with vanilla token generation, it is embarrassingly simple to implement, and it does not require any fine-tuning nor prompt engineering. We implement our model on Llama 3 8B-Instruct and Ministral-8B-Instruct and conduct a qualitative and a quantitative analysis of counterfactually generated text. We conclude with a demonstrative application of counterfactual token generation for bias detection, unveiling interesting insights about the model of the world constructed by large language models.

For agents

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

get_harvested_code_for_paper("2409.17027")
get_code_for_paper("2409.17027")
have("2409.17027")

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

curl https://syntology.ai/api/ran/2409.17027.json

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

[![Syntology run record](https://syntology.ai/api/ran/2409.17027.svg)](https://syntology.ai/paper/2409.17027)

Connect an agent — have() is free.