© 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. 15035
    Prev
    Next
    INDEX
    Explanations

    public class declaration

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

    The neuron is strongly activated by long runs of whitespace indentation (i.e., blocks of repeated space characters).

    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
     auténtico
    -1.02
    少的
    -1.01
     украшения
    -0.98
    hæ
    -0.98
     staden
    -0.96
     hvid
    -0.95
    分で
    -0.94
    ('*
    -0.93
    尔斯
    -0.93
    äv
    -0.93
    POSITIVE LOGITS
     бывает
    1.01
    .
    0.97
    /
    0.96
     /
    0.91
     They
    0.90
     a
    0.89
    (
    0.88
     an
    0.87
    $
    0.85
    '
    0.84
    Activations Density 0.002%

    No Known Activations