Synthetic Model for SAE Training

This repository contains a SyntheticModel for use with SAELens.

Model Info

Number of features: 16,384
Hidden dimension: 512
Hierarchy: Yes
- Root nodes: 128
- Total nodes: 10880
- Max depth: 4
Feature correlation: Yes (scale 0.1)

Usage

from sae_lens.synthetic import SyntheticModel

model = SyntheticModel.from_pretrained("chanind/synth-sae-bench-variations", model_path="superposition/d-512")

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support