© 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. 34-GEMMASCOPE-RES-131K
    7. 31460
    Prev
    Next
    INDEX
    Explanations

    numbers and codes following them

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

    The neuron principally activates on numeric tokens or multi‐digit sequences (e.g. years, dates, codes, phone numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     STUART
    -0.73
    muer
    -0.72
     filed
    -0.71
    鍚
    -0.71
     wak
    -0.71
     Ansicht
    -0.69
    掛けて
    -0.69
     vů
    -0.69
     wzi
    -0.69
    
    -0.69
    POSITIVE LOGITS
    ÁT
    0.82
     Marianne
    0.82
    Moor
    0.79
     麻
    0.78
     silla
    0.75
     Cabo
    0.75
     dumplings
    0.75
    Evange
    0.74
    الب
    0.73
     Curves
    0.73
    Activations Density 0.063%

    No Known Activations