© 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. 129305
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric expressions and quantities (e.g., digits, measurements, decimals) in the text.

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

    qualities and states

    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
    ksikon
    -0.92
    érieur
    -0.87
    positorio
    -0.86
    grazie
    -0.85
    astify
    -0.84
     circadian
    -0.83
     menjalankan
    -0.83
    schedulers
    -0.83
    otient
    -0.82
    thargy
    -0.82
    POSITIVE LOGITS
    そもそも
    0.75
     volume
    0.75
     Bundesrepublik
    0.75
     personnel
    0.72
     Penge
    0.71
    地点
    0.70
    厲害
    0.70
     when
    0.70
     quantity
    0.69
     altogether
    0.69
    Activations Density 0.119%

    No Known Activations