© 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. 72276
    Prev
    Next
    INDEX
    Explanations

    ET CAL HOU NA Corp

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

    The neuron is spotting long runs of whitespace or indentation (i.e. tokens made up of spaces/tabs).

    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
     mentre
    -0.99
    styrelsen
    -0.98
     dessa
    -0.92
     piña
    -0.91
    BEY
    -0.88
     invita
    -0.88
     barbacoa
    -0.86
     -.
    -0.86
     podem
    -0.85
    注定
    -0.85
    POSITIVE LOGITS
     przycis
    1.12
     and
    1.12
     bílá
    1.11
     spodnie
    1.10
    dịch
    1.07
    tôi
    1.05
     odpově
    1.05
     suspendu
    1.04
     modrá
    1.04
    vimos
    1.02
    Activations Density 0.001%

    No Known Activations