© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 114007
    Prev
    Next
    INDEX
    Explanations

    deliberate, delirium, debilitation

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

    The neuron reliably fires on tokens that are part of Latin‐derived word endings (e.g. “–ated,” “–ately,” “–ation,” “–ative,” “–ate,” “–ium”).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     ersten
    -2.44
     \"
    -2.36
     estímulo
    -2.34
     gän
    -2.31
     wesentlichen
    -2.28
    蟜
    -2.28
     sorgfäl
    -2.27
     berücksich
    -2.16
     \'
    -2.14
     entwick
    -2.11
    POSITIVE LOGITS
    涐
    2.44
    茑
    2.41
    ܃
    2.28
    珢
    2.20
    ጴ
    2.19
    dération
    2.11
    猡
    2.06
     angeles
    1.97
     щодо
    1.96
     土産
    1.95
    Activations Density 0.002%

    No Known Activations