© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 127741
    Prev
    Next
    INDEX
    Explanations

    once in a lifetime

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

    The neuron detects the multiword cliché “once-in-a-lifetime” (and its variants like “once-in-a-generation”), flagging rare or unique opportunity expressions.

    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
    \
    -1.87
     yds
    -1.70
    kunden
    -1.50
     itp
    -1.46
     limpar
    -1.45
     Danh
    -1.45
     сказала
    -1.40
     جديد
    -1.38
     kuchen
    -1.37
     ★★★★
    -1.37
    POSITIVE LOGITS
     actually
    1.62
    </h4>
    1.61
     mammoth
    1.52
     که
    1.46
    Spesifikasi
    1.41
    {
    
    
    1.41
    屋外
    1.41
    usky
    1.41
     (
    1.39
     また
    1.39
    Activations Density 0.009%

    No Known Activations