Single SAEs trained on the residual stream activation vectors from every transformer layer simultaneously: https://arxiv.org/abs/2409.04185
Tim Lawson
tim-lawson
AI & ML interests
Mechanistic interpretability, language modelling, semantics
Recent Activity
updated
a model
about 12 hours ago
tim-lawson/sae-pythia-160m-deduped-x64-k32-layers-11
updated
a model
about 12 hours ago
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-11
updated
a model
about 12 hours ago
tim-lawson/sae-pythia-70m-deduped-x64-k32-layers-5
Organizations
None yet
models
191
tim-lawson/sae-pythia-160m-deduped-x64-k32-layers-11
Updated
•
8
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-11
Updated
•
10
tim-lawson/sae-pythia-70m-deduped-x64-k32-layers-5
Updated
•
18
tim-lawson/sae-pythia-70m-deduped-x64-k32-tfm-layers-5
Updated
•
46
tim-lawson/sae-pythia-410m-deduped-x64-k32-layers-19
Updated
tim-lawson/sae-pythia-410m-deduped-x64-k32-tfm-layers-19
Updated
tim-lawson/sae-pythia-410m-deduped-x64-k32-layers-6
Updated
tim-lawson/sae-pythia-410m-deduped-x64-k32-tfm-layers-6
Updated
tim-lawson/sae-pythia-410m-deduped-x64-k32-layers-15
Updated
tim-lawson/sae-pythia-410m-deduped-x64-k32-tfm-layers-15
Updated
datasets
58
tim-lawson/mlsae-gpt2-x64-k32-dists
Preview
•
Updated
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-11-dists
Viewer
•
Updated
•
49.2k
•
10
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-10-dists
Viewer
•
Updated
•
49.2k
•
10
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-8-dists
Viewer
•
Updated
•
49.2k
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-9-dists
Viewer
•
Updated
•
49.2k
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-7-dists
Viewer
•
Updated
•
49.2k
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-5-dists
Viewer
•
Updated
•
49.2k
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-6-dists
Viewer
•
Updated
•
49.2k
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-2-dists
Viewer
•
Updated
•
49.2k
•
9
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-0-dists
Viewer
•
Updated
•
49.2k
•
9