© 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. 15995
    Prev
    Next
    INDEX
    Explanations

    probability fractions with factorials

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

    The neuron strongly activates on numeric tokens and mathematical number expressions (digits, fractions, counts, etc.).

    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
     allerede
    -1.20
    значение
    -1.08
    Romain
    -1.01
     desactivar
    -1.00
     alterar
    -0.98
     красота
    -0.97
    お待ちしております
    -0.96
     ویکی‌پدیای
    -0.95
    iklet
    -0.94
    rages
    -0.93
    POSITIVE LOGITS
     刺繍
    1.03
    おはよう
    1.02
    </em>
    0.99
     робити
    0.97
     beträgt
    0.94
     деца
    0.94
    abonnement
    0.93
     blanches
    0.93
     fewer
    0.92
     barnen
    0.91
    Activations Density 0.019%

    No Known Activations