Papers › The elephant in the interpretability room: Why use attention as explanation when we...

The elephant in the interpretability room: Why use attention as explanation when we have saliency methods?

12 Oct 2020EMNLP (BlackboxNLP) 2020 11arXiv:2010.05607archive 2025-07-28

Jasmijn Bastings, Katja Filippova

There is a recent surge of interest in using attention as explanation of model predictions, with mixed evidence on whether attention can be used as such. While attention conveniently gives us one weight per input token and is easily extracted, it is often unclear toward what goal it is used as explanation. We find that often that goal, whether explicitly stated or not, is to find out what input tokens are the most relevant to a prediction, and that the implied user for the explanation is a model developer. For this goal and user, we argue that input saliency methods are better suited, and that there are no compelling reasons to use attention, despite the coincidence that it provides a weight for each input. With this position paper, we hope to shift some of the recent focus on attention to saliency methods, and for authors to clearly state the goal and user for their explanations.

PaperPDFConference PDFCodeCode 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 16 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 16 samples from 1 repository, 11 ran (5 with no instrument failure; 6 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 between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 11 samples that ran has no recorded date.

jessevig/bertviz mentioned on GitHubpytorchApache-2.0Syntology: 16 samples, 11 ran (6 instrument-failed) · 5 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

16 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
6ran · our draft was wrong
4ran
5unverified

Licence: 2 of the 16 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from jessevig/bertviz. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

jessevig/bertviz community (archive-listed) 16 samples, 11 ran (6 instrument-failed) · 5 unverified
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
bertviz/transformers_neuron_view/modeling_bert.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 0f786c407fb1ee4c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
build_tf_to_pytorch_map in 14 papers bertviz/transformers_neuron_view/modeling_transfo_xl.py ran · our draft was wrong Apache-2.0 (permissive) · 35ef111fc00c21b4 · report fixture: signature model:callable/0/?, config:Config/0/? · returned dict · level recorded 2026-09-22
build_tf_xlnet_to_pytorch_map in 6 papers bertviz/transformers_neuron_view/modeling_xlnet.py ran Apache-2.0 (permissive) · c92ba449e48ecffe · report fixture: signature model:callable/0/?, config:Config/0/?, tf_weights:dict/0/? · returned dict · level recorded 2026-09-25
format_special_chars bertviz/neuron_view.py ran · our draft was wrong Apache-2.0 (permissive) · 11442089a6433797 · report fixture: signature tokens:list/1/str · returned list · level recorded 2026-09-22
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
bertviz/transformers_neuron_view/modeling_bert.py ran · honoured contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · fdc64f4c72036ae4 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
get_attention in 13 papers bertviz/neuron_view.py ran Apache-2.0 (permissive) · d4eae117e16a3f41 · report fixture: signature model:callable/0/?, model_type:str/0/str, tokenizer:MinimalTokenizer/0/?, sentence_a:str/0/str, sentence_b:NoneType/0/?, include_queries_and_keys:bool/0/bool · returned dict · level recorded 2026-09-25
get_masks in 23 papers bertviz/transformers_neuron_view/modeling_xlm.py ran Apache-2.0 (permissive) · f5d8d44f65a08b76 · report fixture: signature slen:int/0/int, lengths:Tensor/1/int, causal:bool/0/bool, padding_mask:NoneType/0/? · returned tuple · level recorded 2026-09-25
load_tf_weights_in_bert in 135 papers bertviz/transformers_neuron_view/modeling_bert.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · 0c145728a35789fa · report fixture: signature model:callable/0/?, config:Config/0/?, tf_checkpoint_path:str/0/str · returned BertModel · level recorded 2026-09-22
prune_conv1d_layer in 20 papers bertviz/transformers_neuron_view/modeling_utils.py ran Apache-2.0 (permissive) · 7183a17dd79327a8 · report fixture: signature layer:callable/0/?, index:Tensor/1/int, dim:int/0/int · returned Conv1D · level recorded 2026-09-25
prune_linear_layer in 40 papers bertviz/transformers_neuron_view/modeling_utils.py ran · our draft was wrong Apache-2.0 (permissive) · d30d6c3098df2c42 · report fixture: signature layer:callable/0/?, index:Tensor/1/int, dim:int/0/int · returned Linear · level recorded 2026-09-22
sample_logits in 16 papers bertviz/transformers_neuron_view/modeling_transfo_xl_utilities.py ran · our draft was wrong Apache-2.0 (permissive) · 93228a3eb5c4d179 · report fixture: signature embedding:callable/0/?, bias:Tensor/1/float, labels:Tensor/2/int, inputs:Tensor/3/float, sampler:LogUniformSampler/0/? · returned Tensor · level recorded 2026-09-22
load_tf_weights_in_gpt2 in 68 papers bertviz/transformers_neuron_view/modeling_gpt2.py unverified Apache-2.0 (permissive) · 654e91efc679fe84 · report
load_tf_weights_in_openai_gpt in 13 papers bertviz/transformers_neuron_view/modeling_openai.py unverified Apache-2.0 (permissive) · 42ee168c3ae12e60 · report
load_tf_weights_in_transfo_xl in 12 papers bertviz/transformers_neuron_view/modeling_transfo_xl.py unverified Apache-2.0 (permissive) · 23d246e59ef9447b · report
load_tf_weights_in_xlnet in 6 papers bertviz/transformers_neuron_view/modeling_xlnet.py unverified Apache-2.0 (permissive) · 4a94662ff3b5c5cf · report
prune_layer in 8 papers bertviz/transformers_neuron_view/modeling_utils.py unverified Apache-2.0 (permissive) · e31093ffc8275cff · report

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