Home › Census › cosine_similarity
cosine_similarity
cosine_similarity: 26 implementations from 30 papers ran on one shared input (census 2026-09-22, battery b986f7e04d79); they produced 11 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:
- Every examined implementation of this name that took an array argument ran on the shared input.
- 9 more implementations from 10 papers took no array argument and ran only on their own fixture arguments; no shared input existed for them, so they are not compared (on 9 the recorded shared-output digest is the own-fixture digest).
- Every compared output was digested.
- The harness also ran torch's own computation of this name under 6 named conventions; those rows are not implementations, are not counted, and only label a cluster whose digest they share.
Bucket 1 of 5: arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32
13 implementations from 16 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 5 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 5 implementations 5 papers adc71adfbba8 recorded values identical same digest as torch: matrix |
model.py 0b4f1ab0 recorded x:2/float/float32, y:2/float/float32
modeling.py 19bb073c recorded x:2/float/float32, y:2/float/float32 models/proto/protaugment.py 3ae294e8 recorded x:2/float/float32, y:2/float/float32 dora/dora.py 810e3d44 recorded a:2/float/float32, b:2/float/float32; eps=1e-05 laft/laft.py 88695493 recorded x1:2/float/float32, x2:2/float/float32; eps=1e-08 |
[0.997778, -0.117586, 0.738272, -0.23741, -0.117586, 0.997491, -0.39831, 0.191101, …] shape [4, 4] · float32 · Tensor |
| 4 implementations 6 papers c914e8188e43 recorded values identical same digest as torch: rowwise |
train_watermark_model.py 2a673044 recorded x:2/float/float32, y:2/float/float32 one code sha held from 2 papers' repositories segmentation/losses/dist_kd.py 38ede58b recorded x:2/float/float32, y:2/float/float32; eps=1e-08
lib/models/losses/dist_kd.py b5f44526 recorded a:2/float/float32, b:2/float/float32; eps=1e-08 evaluation/caculate_R_precison.py e0c14774 recorded x1:2/float/float32, x2:2/float/float32; dim=1, eps=1e-08 |
[1, 1, 1, 1] shape [4] · float32 · Tensor |
| 2 implementations 3 papers ff59a1024481 recorded values identical |
train_dit.py e157ae27 recorded ta:2/float/float32, tb:2/float/float32 src/analysis/cluster_analysis.py bcb8d066 recorded batch1:2/float/float32, batch2:2/float/float32; eps=1e-08 |
[1.00054, -0.119116, 0.739214, -0.239589, -0.119116, 1.00397, -0.400068, 0.192469, …] shape [4, 4] · float32 · Tensor |
| 1 implementation 1 paper 13169b86a884 |
evaluation/score_single.py 892eedee recorded vec1:2/float/float32, vec2:2/float/float32 |
[1, -0.118974, 0.739129, -0.239338, -0.118974, 1, -0.399015, 0.191531, …] shape [4, 4] · float32 · ndarray |
| 1 implementation 1 paper e331ee3b84bf |
main_lip.py 2be8edc2 recorded grad1:2/float/float32, grad2:2/float/float32 |
[32.4175] shape [] · float32 · Tensor |
Bucket 2 of 5: arg 1: rank 1, kind float, dtype float64 · arg 2: rank 1, kind float, dtype float64
9 implementations from 10 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 3 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 7 implementations 8 papers 6c3c396ed6b5 recorded values differ by up to 2.22e-16 |
sentence_chunk.py 91467442 recorded vec1:1/float/float64, vec2:1/float/float64 ReSo/agent_graph/agent_graph.py 07e648de recorded vec1:1/float/float64, vec2:1/float/float64 explanatory_contextual_retrieval.py 273c1c7b recorded vector1:1/float/float64, vector2:1/float/float64 scripts/experiment.py 440007de recorded vec_a:1/float/float64, vec_b:1/float/float64 infer/infer_gpt.py 729c43fa recorded a:1/float/float64, b:1/float/float64 utils/metrics.py ac76b4e9 recorded x:1/float/float64, y:1/float/float64 compute_mexa.py dbe1b491 recorded array1:1/float/float64, array2:1/float/float64 |
[1] shape [] · float64 |
| 1 implementation 1 paper 96a52f1bd609 |
optex_jax/OptEx_JAX.py 99b2c601 recorded vector1:1/float/float64, vector2:1/float/float64 |
[0.999998] shape [] · float64 |
| 1 implementation 1 paper af5570f5a181 |
poisoning/attack_eval_pipeline.py 1eb9d456 recorded a:1/float/float64, b:1/float/float64 |
[0] shape [] · float64 |
Bucket 3 of 5: arg 1: rank 1, kind float, dtype float32 · arg 2: rank 1, 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); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 2 implementations 2 papers 6c3c396ed6b5 recorded values identical same digest as torch: matrix, rowwise |
four_layer_convergence.py 6dc29ca9 recorded x1:1/float/float32, x2:1/float/float32 src/losses.py e8a410a4 recorded w1:1/float/float32, w2:1/float/float32 |
[1] shape [] · float64 · float |
Bucket 4 of 5: arg 1: rank 3, kind float, dtype float32 · arg 2: rank 3, 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) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper d669d7c421a3 |
utils/proto_utils.py e0ea79d7 recorded x:3/float/float32, y:3/float/float32 |
[1, 0.4665, -0.660587, -0.0709421, 0.4665, 1, -0.690947, -0.204146, …] shape [2, 4, 4] · float32 · Tensor |
Bucket 5 of 5: arg 1: rank 4, kind float, dtype float32 · arg 2: 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) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper acfc7c36fce5 |
models.py 6d870362 recorded x:4/float/float32, y:4/float/float32; patch_size=3 |
[1, 1, 1, 1, 1, 1, 1, 1, …] shape [2, 4, 4] · 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/cosine-similarity.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