© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 42126
    Prev
    Next
    INDEX
    Explanations

    date and time markers

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

    The neuron fires on numeric tokens and number-containing expressions (dates, measured values, IDs, percentages, etc.).

    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
     either
    -1.66
     to
    -1.59
    向了
    -1.55
     How
    -1.54
     nasıl
    -1.52
    んでいた
    -1.48
     how
    -1.39
     хочет
    -1.36
     compleja
    -1.35
    ৹
    -1.35
    POSITIVE LOGITS
    ’
    1.48
     dévelo
    1.45
     brillant
    1.41
    此
    1.38
    『
    1.38
    "
    1.34
    ✔️
    1.33
     églises
    1.32
     poë
    1.30
     fabrik
    1.29
    Activations Density 0.045%

    No Known Activations