Home › Census › PositionalEncoding
PositionalEncoding
PositionalEncoding: 13 implementations from 9 papers ran on one shared input (census 2026-09-22, battery b986f7e04d79); they produced 9 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:
- 15 more implementations of this name were examined but did not run on the shared input (RuntimeError 15).
- 2 more implementations from 4 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 2 the recorded shared-output digest is the own-fixture digest).
- Every compared output was digested.
- 13 compared members are class-bearing (tagged class): the class was constructed with seeded weights in evaluation mode and then called, so its recorded output depends on that initialisation as well as on the forward computation.
Bucket 1 of 5: arg 1: rank 3, kind float, dtype float32
9 implementations from 6 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 |
|---|---|---|
| 3 implementations 2 papers 10d8865d2703 recorded values differ by up to 1.19e-07 |
class model/standard.py 4f5b9488 recorded x:3/float/float32 class models.py 5dc71b21 recorded x:3/float/float32 class model/scratch_model.py 92aea12e recorded x:3/float/float32 |
[-0.893127, 2.68285, -1.2588, 0.741794, 0.150996, -0.434759, -0.340146, 0.618874, …] shape [2, 4, 8] · float32 · Tensor |
| 2 implementations 2 papers 28cd979073fb recorded values identical |
class lib/eeg_transformer.py 429a0628 recorded x:3/float/float32 class models/TimeDART.py ba3da553 recorded x:3/float/float32 |
[-0, 2.98095, -1.39867, 0, 0.167773, -0.483066, -0.37794, 0.687638, …] shape [2, 4, 8] · float32 · Tensor |
| 2 implementations 2 papers f52b65ca7dca recorded values identical |
class uniMASK/transformer.py 117a93cf recorded x:3/float/float32 class model.py b9af059f recorded x:3/float/float32 |
[-0, 2.98095, -1.39867, 0, 0.167773, -0.483066, -0.37794, 0.687638, …] shape [2, 4, 8] · float32 · Tensor |
| 1 implementation 1 paper 766cc7234ad6 |
class grounded_compgen_research/models/imitation/conv_transformer.py 6d88134a recorded x:3/float/float32 |
[0, 1, 0, 1, 0.841471, 0.540302, 0.00999983, 0.99995] shape [2, 1, 4] · float32 · Tensor |
| 1 implementation 1 paper f5efe631655a |
class model.py 6ae60d8c recorded x:3/float/float32 |
[-0, 1.98095, -1.39867, -0, 0.167773, -1.48307, -0.37794, -0.312362, …] shape [2, 4, 8] · float32 · Tensor |
Bucket 2 of 5: arg 1: rank 1, 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 a53e4536a65f |
class model.py 078a3713 recorded noise_level:1/float/float32 |
[0.48179, 0.48179, 0.48179, 0.48179, 0.876287, 0.876287, 0.876287, 0.876287, …] shape [8, 8] · float32 · Tensor |
Bucket 3 of 5: arg 1: rank 1, 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) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper bea48acc7c12 |
class k-edge-identification/src/ehnn/models.py 2fd498d9 recorded x:1/int/int64 |
[0.909297, -0.416147, 0.0199987, 0.9998, 0.841471, 0.540302, 0.00999983, 0.99995, …] shape [8, 4] · float32 · Tensor |
Bucket 4 of 5: arg 1: rank 2, 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 ac63a8474c76 |
class golf/model.py a043ead7 recorded coords:2/float/float32; rays=None |
[-0.581084, -0.945823, -0.614184, 0.969382, 0.476075, -0.837326, -0.91554, 0.736511, …] shape [4, 128] · float32 · Tensor |
Bucket 5 of 5: arg 1: rank 2, 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) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper a721946fbb2c |
class transformer/models.py c1e9ee7b recorded inputs:2/int/int64 |
[0.909297, -0.416147, 0.0199987, 0.9998, 0.14112, -0.989992, 0.0299955, 0.99955, …] shape [4, 8, 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/positionalencoding.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