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
6 minutes ago
tim-lawson/temp-pythia-70m-deduped-x64-k32-l3
updated
a model
1 day ago
tim-lawson/temp-pythia-70m-deduped-x256-k32-l3
published
a model
1 day ago
tim-lawson/temp-pythia-70m-deduped-x256-k32-l3
Organizations
None yet
Collections
6
Papers
1
models
285
tim-lawson/temp-pythia-70m-deduped-x64-k32-l3
Updated
tim-lawson/temp-pythia-70m-deduped-x256-k32-l3
Updated
tim-lawson/temp-pythia-70m-deduped-x128-k32-l3
Updated
tim-lawson/temp-pythia-410m-deduped-x64-k32-l9
Updated
tim-lawson/temp-pythia-70m-deduped-x64-k32-l3-j10.0
Updated
tim-lawson/temp-pythia-70m-deduped-x64-k32-l3-j0.05
Updated
tim-lawson/temp-pythia-160m-deduped-x64-k32-l7-j500.0
Updated
tim-lawson/temp-pythia-70m-deduped-x64-k32-l3-j0.1
Updated
tim-lawson/temp-pythia-70m-deduped-x2-k32-l3
Updated
tim-lawson/temp-pythia-70m-deduped-x64-k32-l3-j0.01
Updated
datasets
60
tim-lawson/mlsae-Llama-3.2-3B-x64-k32-dists
Viewer
•
Updated
•
197k
•
45
tim-lawson/mlsae-gemma-2-2b-x64-k32-dists
Viewer
•
Updated
•
147k
•
42
tim-lawson/mlsae-gpt2-x64-k32-dists
Preview
•
Updated
•
38
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-11-dists
Viewer
•
Updated
•
49.2k
•
39
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-10-dists
Viewer
•
Updated
•
49.2k
•
39
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-8-dists
Viewer
•
Updated
•
49.2k
•
39
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-9-dists
Viewer
•
Updated
•
49.2k
•
40
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-7-dists
Viewer
•
Updated
•
49.2k
•
40
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-5-dists
Viewer
•
Updated
•
49.2k
•
40
tim-lawson/sae-pythia-160m-deduped-x64-k32-tfm-layers-6-dists
Viewer
•
Updated
•
49.2k
•
40