© 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. 129404
    Prev
    Next
    INDEX
    Explanations

    This neuron fires strongly on content-bearing (informative) words (i.e. non‐function words or “keywords”).

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

    common phrases and relationships

    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
     zowel
    -1.73
     célè
    -1.62
     spektak
    -1.55
     énorm
    -1.55
     wondrous
    -1.52
     acheter
    -1.48
     både
    -1.47
    𓆏
    -1.46
     vijf
    -1.46
     rése
    -1.46
    POSITIVE LOGITS
     can
    1.71
     it
    1.61
     Not
    1.60
     if
    1.59
     In
    1.57
     could
    1.55
     from
    1.55
     k
    1.55
     There
    1.54
     One
    1.53
    Activations Density 0.142%

    No Known Activations