© 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. 128530
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on multi‐digit numeric tokens (i.e. on answer numbers rather than on words).

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

    mathematical problems and phrases

    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
    ofes
    -0.87
     EC
    -0.87
     какого
    -0.81
    EJB
    -0.81
     days
    -0.79
    Maryland
    -0.78
    アンド
    -0.77
    estay
    -0.77
     Maryland
    -0.75
     promptly
    -0.75
    POSITIVE LOGITS
     experimenta
    0.91
    UNDLE
    0.91
    Lato
    0.85
    ilities
    0.84
    ISTRO
    0.84
    ності
    0.81
     declarat
    0.80
     guardia
    0.79
    ;&#
    0.78
     esponja
    0.76
    Activations Density 0.027%

    No Known Activations