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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111186
    Prev
    Next
    INDEX
    Explanations

    return to navigation

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

    The neuron detects navigation or “go back” style link or button labels (e.g. “Back to,” “Return to,” “Go back to,” “Home,” “Previous”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     you
    -1.70
     a
    -1.55
    ඇ
    -1.41
    ыр
    -1.34
    devils
    -1.33
     its
    -1.30
    васто
    -1.28
    ٞ
    -1.25
    いしい
    -1.25
     उसने
    -1.23
    POSITIVE LOGITS
    begin
    1.63
    𔘓
    1.48
    CHREIBUNG
    1.41
     Họ
    1.41
     शुरु
    1.39
     mußte
    1.34
    end
    1.33
     etmeye
    1.33
    放入
    1.30
     juzgar
    1.30
    Activations Density 0.032%

    No Known Activations