You can deploy this model and start earning money today!

decoderesearch/
synth-sae-bench-16k-v1

Number of features: 16,384
Hidden dimension: 768
Hierarchy: Yes Root nodes: 42 Total nodes: 16384 Max depth: 3
Feature correlation: Yes (scale 0.1)

Views

No views yet

Synthetic Model for SAE Training

This repository contains a SyntheticModel for use with SAELens.

python

from sae_lens.synthetic import SyntheticModel

model = SyntheticModel.from_pretrained("decoderesearch/synth-sae-bench-16k-v1")