© 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. 73544
    Prev
    Next
    INDEX
    Explanations

    years and numbers

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

    This neuron detects the sequence of a numeric year immediately followed by the words “to be” (e.g. “2007 to be”).

    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
     some
    -1.77
     個人
    -1.45
    －－
    -1.45
     abundantly
    -1.42
     still
    -1.41
    簡介
    -1.41
    On
    -1.41
    To
    -1.40
    At
    -1.40
    If
    -1.38
    POSITIVE LOGITS
    ↵
    1.41
     olie
    1.40
     CHRIS
    1.33
     aldus
    1.33
     zand
    1.33
    2
    1.32
     zout
    1.31
    鹇
    1.28
    Kelebihan
    1.26
    f
    1.23
    Activations Density 0.018%

    No Known Activations