© 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. 22-GEMMASCOPE-RES-131K
    7. 14975
    Prev
    Next
    INDEX
    Explanations

    specifying variables after where

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

    The neuron fires on numeric tokens—digits or numbers (e.g. equation numbers, constants, list indices) appearing in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     accompagner
    -1.09
    点を
    -0.99
     nivelul
    -0.91
    ౖ
    -0.90
     парфю
    -0.84
    ⏑
    -0.84
    Demographics
    -0.84
    nhi
    -0.82
     specif
    -0.82
     プル
    -0.81
    POSITIVE LOGITS
     TWO
    1.19
     because
    1.16
     for
    1.16
     впечат
    0.91
    Despite
    0.91
     AWESOME
    0.90
    科
    0.89
    aiuto
    0.89
    umahan
    0.88
    臧
    0.88
    Activations Density 0.023%

    No Known Activations