© 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. 85425
    Prev
    Next
    INDEX
    Explanations

    names: Janet, Angus, Bernstein, Becker

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

    The neuron chiefly activates on multi‐digit numeric tokens—especially larger numbers like years, IDs, or measurements.

    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
    我的
    -2.70
    os
    -2.53
    ti
    -2.38
    u
    -2.30
    t
    -2.30
     Which
    -2.25
     Mitarbeitern
    -2.25
    j
    -2.22
    la
    -2.20
    ction
    -2.20
    POSITIVE LOGITS
    虽说
    2.05
     способ
    2.05
    ſted
    1.88
     prêtre
    1.84
    anglès
    1.83
     paraly
    1.81
     のみ
    1.80
     sobą
    1.73
    町の
    1.72
     kadang
    1.72
    Activations Density 0.004%

    No Known Activations