© 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. 73146
    Prev
    Next
    INDEX
    Explanations

    numbers and lists

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

    The neuron primarily activates on numeric characters, i.e. digits in numeric literals or long number sequences.

    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
    reactstrap
    -0.94
     there
    -0.86
    UARIO
    -0.84
    どんどん
    -0.82
    ៩
    -0.82
    weird
    -0.79
     различни
    -0.79
    ucc
    -0.78
    淡淡
    -0.77
     хора
    -0.75
    POSITIVE LOGITS
    jada
    1.00
    מא
    0.94
     hieronder
    0.90
    的时候
    0.88
    攵
    0.87
    zony
    0.87
    AddField
    0.86
     doveva
    0.86
     strö
    0.85
    ofada
    0.84
    Activations Density 0.004%

    No Known Activations