Home › Census › attention

attention

Syntologyfunction-name censuscensus 2026-09-22battery b986f7e04d79all samples with this name

attention: 11 implementations from 10 papers ran on one shared input (census 2026-09-22, battery b986f7e04d79); they produced 7 distinct outputs across 5 buckets, one shared input per bucket.

Identical values to six decimals (the recorded digest) on the shared input are agreement on those inputs, not equivalence. Implementations are compared only within one bucket, the positional (rank, kind, dtype) of each array argument; the argument name is not part of the key because the harness draws the shared array from (rank, kind) and casts it to the dtype, whatever the name; each bucket's shared input is fixed by that key, so members of one bucket saw bitwise-identical inputs under their own scalar arguments. A cluster is the set of members whose recorded output digest is identical. Nothing here says which computation a paper's method intended, and nothing reproduces a paper's results.

Not compared, and not in the tables or the counts above:

Bucket 1 of 5: arg 1: rank 4, kind float, dtype float32 · arg 2: rank 4, kind float, dtype float32 · arg 3: rank 4, kind float, dtype float32 · arg 4: rank 4, kind float, dtype float32

4 implementations from 3 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 2 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)MembersShared output on this bucket's input
3 implementations
2 papers
c190e723cd2e
recorded values identical

pykt/models/simplekt.py 5bd62b87
recorded q:4/float/float32, k:4/float/float32, v:4/float/float32, mask:4/float/float32; d_k=5, dropout=Dropout(p=0.1, inplace=False), zero_pad=False

models/pakt.py 72814a15
recorded q:4/float/float32, k:4/float/float32, v:4/float/float32, mask:4/float/float32; d_k=5, dropout=Dropout(p=0.1, inplace=False), zero_pad=False

pykt/models/simplekt.py e05cd8a1
recorded q:4/float/float32, k:4/float/float32, v:4/float/float32, mask:4/float/float32; d_k=5, dropout=Dropout(p=0.1, inplace=False), zero_pad=False
[0.260109, -0.0500833, -0.129173, 0.174312, 0.580391, -0.0257366, -0.549415, 0.807632, …]
shape [2, 3, 4, 4] · float32 · Tensor
1 implementation
1 paper
ff86505cf778

model/TransKT.py 134e95d1
recorded q:4/float/float32, k:4/float/float32, v:4/float/float32, mask:4/float/float32; d_k=5, dropout=Dropout(p=0.1, inplace=False), zero_pad=True
[0, 0, 0, 0, 0.580391, -0.0257366, -0.549415, 0.807632, …]
shape [2, 3, 4, 4] · float32 · Tensor

Bucket 2 of 5: arg 1: rank 4, kind float, dtype float32

3 implementations from 3 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)MembersShared output on this bucket's input
3 implementations
3 papers
42c78be7a46f
recorded values identical
one code sha held from 2 papers' repositories
unlearn.py b817aa66
recorded x:4/float/float32

Our_code/Pytorch/zero-shot-baseline.py 8031f2b6
recorded x:4/float/float32

Our_code/Pytorch/Basic_experiments/exp-16_2-16_1.py ab5d18fb
recorded x:4/float/float32
[0.0641893, 0.105176, 0.189701, 0.2541, 0.159976, 0.375468, 0.0196281, 0.29745, …]
shape [2, 16] · float32 · Tensor

Bucket 3 of 5: arg 1: rank 3, kind float, dtype float32 · arg 2: rank 3, kind float, dtype float32 · arg 3: rank 3, kind float, dtype float32

2 implementations from 2 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 2 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)MembersShared output on this bucket's input
1 implementation
1 paper
02617d7c9917

utils.py 72ce5bea
recorded Q:3/float/float32, K:3/float/float32, V:3/float/float32
[-1.00379, 1.34099, -1.1562, -0.339274, 0.212062, -1.08009, -0.280365, -0.139151, …]
shape [2, 4, 8] · float32 · Tensor
1 implementation
1 paper
c09ff4f743dd

kandinsky2/model/text2im_model2_1.py fe304798
recorded q:3/float/float32, k:3/float/float32, v:3/float/float32; d_k=4
[-0.992026, 1.49343, -1.20681, -0.310881, 0.204297, -1.2598, -0.297142, -0.264622, …]
shape [2, 4, 8] · float32 · Tensor

Bucket 4 of 5: arg 1: rank 4, kind float, dtype float32 · arg 2: rank 4, kind float, dtype float32 · arg 3: rank 4, kind float, dtype float32 · arg 4: rank 4, kind int, dtype int64

1 implementation from 1 paper share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)MembersShared output on this bucket's input
1 implementation
1 paper
cc2e4b612f8d

models/oclr.py 86341ad6
recorded q:4/float/float32, k:4/float/float32, v:4/float/float32, mask:4/int/int64; d_k=4, dropout=Dropout(p=0.1, inplace=False)
[0.0193632, -0.00746569, -0.0671406, -0.000716995, 0.652213, 0.00248403, -0.340817, 1.00866, …]
shape [1, 1, 2, 3, 4, 4] · float32 · Tensor

Bucket 5 of 5: arg 1: rank 4, kind float, dtype float32 · arg 2: rank 4, kind float, dtype float32 · arg 3: rank 4, kind float, dtype float32

1 implementation from 1 paper share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.

Cluster (same digest to six decimals)MembersShared output on this bucket's input
1 implementation
1 paper
ac07ca26d031

hyvideo/modules/models.py 7f6ea851
recorded q:4/float/float32, k:4/float/float32, v:4/float/float32; mode='vanilla', drop_rate=0.0, attn_mask=None, causal=False, cu_seqlens_q=None, cu_seqlens_kv=None, max_seqlen_q=None, max_seqlen_kv=None, batch_size=2
[0.571789, 0.644474, 1.26234, 0.830062, 0.511626, 0.321406, -0.267345, 0.83323, …]
shape [2, 3, 16] · float32 · Tensor

Identical values to six decimals (the recorded digest) on the shared input are agreement on those inputs, not equivalence; where a cluster's members carry recorded values, the largest difference among them is shown under the cluster. Paper titles are the archive's archive 2025-07-28 where the paper is in the archive and the graph's where it was added by Syntology; papers with no page here are shown by their recorded paper id only. A paper count above the implementation count means one implementation (one code sha) is held from several papers' repositories and counts once. Per-sample status, licence and fingerprint records for each paper are on its paper page. JSON twin: /census/attention-2.json.

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