Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Docs
  • Enterprise
  • Pricing

  • Log In
  • Sign Up
science-of-finetuning 's Collections
Chat-data datasets
BatchTopK gemma-2-2b chat SAEs
Base-Chat middle layer Crosscoders

Base-Chat middle layer Crosscoders

updated Jun 3

A collection of crosscoders trained for our paper "Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning" https://arxiv.org/abs/2504.0

Upvote
-

  • science-of-finetuning/gemma-2-2b-crosscoder-l13-mu4.1e-02-lr1e-04

    Feature Extraction • 0.7B • Updated Nov 22, 2024 • 12 • 4

  • science-of-finetuning/gemma-2-2b-L13-k100-lr1e-04-local-shuffling-CCLoss

    0.7B • Updated Mar 5 • 3

  • science-of-finetuning/Llama-3.2-1B-L8-mu3.6e-02-lr1e-04-local-shuffling-CrosscoderLoss

    0.5B • Updated May 15 • 6

  • science-of-finetuning/Llama-3.2-1B-L8-k100-lr1e-04-local-shuffling-Crosscoder

    0.5B • Updated May 15 • 1

  • science-of-finetuning/Meta-Llama-3.1-8B-L16-k200-lr1e-04-local-shuffling-Crosscoder-ni0.3-ka1k5k

    2B • Updated May 21 • 2

  • science-of-finetuning/Meta-Llama-3.1-8B-L16-mu2.1e-02-lr1e-04-local-shuffling-CrosscoderLoss

    2B • Updated May 21 • 5
Upvote
-
  • Collection guide
  • Browse collections
Company
TOS Privacy About Jobs
Website
Models Datasets Spaces Pricing Docs