© 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. 33749
    Prev
    Next
    INDEX
    Explanations

    solving equations

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

    The neuron fires when it sees numeric constants (digits or numbers) in mathematical expressions.

    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
     realizaron
    -0.98
    競技
    -0.94
    ILDREN
    -0.91
    
    -0.89
     påvir
    -0.85
     Actions
    -0.84
     Glaubens
    -0.83
     Handlungen
    -0.83
    Zitat
    -0.83
     administrativos
    -0.82
    POSITIVE LOGITS
     ump
    0.91
    llbracket
    0.82
     frater
    0.82
    format
    0.81
     indetermin
    0.79
     raving
    0.78
     summarizes
    0.77
     what
    0.77
     summarised
    0.77
     summarize
    0.76
    Activations Density 0.007%

    No Known Activations