© 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. 129110
    Prev
    Next
    INDEX
    Explanations

    This neuron responds to numeric and other non-alphabetic tokens (e.g. numbers and isolated hyphen/soft-hyphen fragments).

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

    sign-age and place names

    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
     áll
    -0.85
     smtplib
    -0.83
     varandra
    -0.82
    اقتصاد
    -0.81
    newLine
    -0.79
    そこは
    -0.79
    思议
    -0.79
     smör
    -0.78
     licor
    -0.78
    ói
    -0.78
    POSITIVE LOGITS
    Ặ
    0.88
    %)$
    0.88
     psu
    0.85
     CMY
    0.83
     dispel
    0.82
    ۂ
    0.80
    innerHTML
    0.80
     eventuali
    0.79
     DIFFERENT
    0.78
    =$?
    0.78
    Activations Density 0.004%

    No Known Activations