© 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. 72272
    Prev
    Next
    INDEX
    Explanations

    code comments and headers

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

    The neuron fires on numeric tokens (sequences of digits), effectively detecting numbers in the text.

    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
     back
    -0.82
     whose
    -0.82
    mathrm
    -0.82
    要点
    -0.82
    滑动
    -0.82
     separate
    -0.81
     abierta
    -0.78
    弹窗
    -0.78
     Schriften
    -0.77
    ту
    -0.76
    POSITIVE LOGITS
     Coro
    0.94
     comité
    0.94
    Carlton
    0.89
     کور
    0.85
    ılık
    0.85
     slår
    0.84
     aiment
    0.84
    ính
    0.84
    UNCIL
    0.83
    tapkan
    0.83
    Activations Density 0.034%

    No Known Activations