© 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. 10-GEMMASCOPE-RES-131K
    7. 58495
    Prev
    Next
    INDEX
    Explanations

    duration until a specific point

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

    This neuron detects tokens that signal temporal durations or extended time periods (e.g. “for,” “long,” “time,” “until,” “years”).

    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
     from
    -1.36
     viviendo
    -1.29
     longterm
    -1.13
    ния
    -1.09
     onları
    -1.09
     Deiner
    -1.03
    从来
    -1.02
     sumamente
    -1.02
    事を
    -1.01
     fallecido
    -1.01
    POSITIVE LOGITS
     広場
    1.29
     enough
    1.23
    錾
    1.13
    BrowserModule
    1.13
    Зачем
    1.12
     části
    1.10
    Качество
    1.10
    szak
    1.10
     休閒
    1.09
    ଵ
    1.09
    Activations Density 0.062%

    No Known Activations