© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128402
    Prev
    Next
    INDEX
    Explanations

    The neuron does not detect any patterns—it remains inactive and never activates for any tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    prefix then suffix

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nother
    -1.34
    
    -1.31
    featuring
    -1.27
    部分が
    -1.23
    部分は
    -1.20
     SPECIAL
    -1.20
    lorn
    -1.20
     konsult
    -1.17
    wouldn
    -1.17
    sitting
    -1.17
    POSITIVE LOGITS
     and
    1.88
     (
    1.80
     from
    1.52
     at
    1.50
     while
    1.48
     before
    1.46
     seems
    1.41
     with
    1.40
     our
    1.39
     or
    1.37
    Activations Density 0.000%

    No Known Activations