© 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. 128296
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens (integers or decimal numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    programming code snippets

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    のだろうか
    -0.81
    ⟹
    -0.80
     teknis
    -0.79
     Used
    -0.79
     gebruikt
    -0.77
     каса
    -0.76
     eorum
    -0.75
    prf
    -0.75
    υσ
    -0.75
    Original
    -0.75
    POSITIVE LOGITS
     ännu
    0.92
     ложки
    0.90
    ！）
    0.88
     can
    0.86
    eles
    0.85
    lés
    0.85
     undeniably
    0.82
    ών
    0.80
     quickly
    0.80
     from
    0.80
    Activations Density 0.029%

    No Known Activations