© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67759
    Prev
    Next
    INDEX
    Explanations

    database syntax

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

    This neuron lights up on unusually rare or domain‐specific tokens (e.g. uncommon words, numbers, or identifiers) regardless of context.

    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
     in
    -2.11
     of
    -1.93
     on
    -1.84
     to
    -1.72
     how
    -1.45
     from
    -1.39
     now
    -1.28
     taken
    -1.23
     you
    -1.22
    いただけます
    -1.22
    POSITIVE LOGITS
    fordshire
    1.45
    ttes
    1.39
    moiselle
    1.28
    fuls
    1.28
     olika
    1.27
    alised
    1.27
    /</
    1.24
     Leverkusen
    1.24
    ciled
    1.24
     sacrific
    1.24
    Activations Density 0.020%

    No Known Activations