© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14661
    Prev
    Next
    INDEX
    Explanations

    scrubbing or scrutinizing

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

    The neuron fires on words beginning with the letters “scru” (e.g. the “scru-” root in words like scrub and scrutiny).

    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
     They
    -2.70
     But
    -2.33
    焄
    -2.28
    ⁛
    -2.28
     다른
    -2.28
    擽
    -2.19
     Responding
    -2.16
     이미
    -2.09
     상품
    -2.08
     Consist
    -2.05
    POSITIVE LOGITS
    '
    2.75
    i
    2.55
    ("
    2.47
     continúas
    2.31
    ’
    2.22
    ="
    2.20
    this
    2.14
    that
    2.11
    ed
    2.02
    ily
    2.02
    Activations Density 0.003%

    No Known Activations