© 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. 65171
    Prev
    Next
    INDEX
    Explanations

    size Female

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

    This neuron fires whenever it sees sequences of digits or numeric data—such as table entries, quantities, or equations.

    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
     The
    -1.31
    ;
    -0.95
     that
    -0.95
    好不好
    -0.90
     arrange
    -0.86
    couts
    -0.85
     secondly
    -0.84
    :
    -0.83
     produkt
    -0.83
     suggest
    -0.82
    POSITIVE LOGITS
     geweldige
    1.11
    久しぶりに
    1.06
    久しぶりの
    1.03
    hình
    1.01
     снова
    1.00
    Zahl
    1.00
     prachtige
    0.99
     magnifique
    0.98
     pengaruh
    0.96
    ământ
    0.94
    Activations Density 0.004%

    No Known Activations