© 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. 22-GEMMASCOPE-RES-131K
    4. 130828
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on scheduling or timing cues—dates, times, days of the week, and related event-schedule terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    time before events

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    '
    -1.12
    const
    -1.07
     بعدی
    -1.03
     häufigsten
    -1.00
     Gästen
    -0.98
    #
    -0.97
    new
    -0.93
     aparecen
    -0.92
    čili
    -0.91
     ihres
    -0.91
    POSITIVE LOGITS
     preceding
    1.87
     before
    1.66
     antes
    1.62
     trước
    1.61
     prior
    1.57
    ก่อน
    1.56
     قبل
    1.52
    Antes
    1.52
     prelude
    1.44
    Sebelum
    1.42
    Activations Density 0.037%

    No Known Activations