© 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. 34305
    Prev
    Next
    INDEX
    Explanations

    search, error, empty, type, efficiency

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

    The neuron fires on tokens containing non-Latin characters (e.g. Chinese logograms or other non-Latin script).

    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
     also
    -1.60
     about
    -1.47
     a
    -1.42
     being
    -1.31
     an
    -1.24
     seeing
    -1.23
    textTheme
    -1.20
     like
    -1.18
     watching
    -1.17
    curated
    -1.16
    POSITIVE LOGITS
    Mvh
    1.30
    Presentación
    1.23
    تفسير
    1.22
    orienti
    1.17
    آموزش
    1.16
    درمان
    1.14
     tandis
    1.14
    Région
    1.08
    ociação
    1.07
     réessayer
    1.06
    Activations Density 0.009%

    No Known Activations