© 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. 34380
    Prev
    Next
    INDEX
    Explanations

    Selective spectro

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects verbs and markers of reported speech or narrative exposition (e.g. “tells,” “informs,” “realizing that,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     kebersihan
    -0.97
    嚐
    -0.96
     many
    -0.95
     several
    -0.93
     on
    -0.91
     and
    -0.90
    тник
    -0.89
     acestui
    -0.88
    ⛷
    -0.88
    分鐘
    -0.87
    POSITIVE LOGITS
     to
    1.49
    škoda
    1.36
    盦
    1.28
    schöne
    1.23
    完成です
    1.21
    EdgeInsets
    1.17
     Semoga
    1.15
    obrázek
    1.15
     Versuche
    1.12
     comunicarse
    1.12
    Activations Density 0.004%

    No Known Activations