© 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. 25659
    Prev
    Next
    INDEX
    Explanations

    code keywords and syntax

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

    The neuron strongly activates on numeric tokens (digits or number literals) appearing anywhere 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
    ьо
    -0.98
    hammed
    -0.94
    ｂ
    -0.92
    дан
    -0.90
     когда
    -0.88
    ないので
    -0.88
    やう
    -0.88
     lorsqu
    -0.86
    oston
    -0.85
    letzt
    -0.85
    POSITIVE LOGITS
    Földrajz
    0.94
     Menschheit
    0.93
    ulite
    0.91
    Requires
    0.87
     carreras
    0.86
     Aman
    0.85
     obé
    0.84
    Törté
    0.82
    requires
    0.82
     fabrique
    0.82
    Activations Density 0.020%

    No Known Activations