© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 18-RES-MATRYOSHKA-DC
    7. 889
    Prev
    Next
    INDEX
    Explanations

    the word "follow" followed by "up," "on," or another word.

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.18.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     BoxFit
    -0.58
    astă
    -0.58
     Greenfield
    -0.57
    shells
    -0.56
     AppModule
    -0.56
     McCarty
    -0.56
     enn
    -0.56
     än
    -0.54
     Hawley
    -0.54
     anfangen
    -0.53
    POSITIVE LOGITS
     Follows
    1.17
     follows
    1.09
     FOLLOW
    1.02
    follow
    1.01
    Followed
    1.01
    follows
    1.00
     follow
    0.98
    Follow
    0.96
     Followed
    0.93
     Follow
    0.92
    Activations Density 0.290%

    No Known Activations