© 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. 33998
    Prev
    Next
    INDEX
    Explanations

    Four Secrets, Orrin, Beep

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

    The neuron is keyed to spotting Unicode escape sequences (e.g. “\uXXXX”) or similar hex‐code fragments in text.

    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
    を
    -1.58
    は
    -1.54
    などが
    -1.54
     japones
    -1.50
     algas
    -1.49
    もあります
    -1.45
    も
    -1.43
    などを
    -1.40
    <bos>
    -1.38
    など
    -1.35
    POSITIVE LOGITS
    這些
    1.54
    為に
    1.52
    __':
    1.51
    _
    1.47
    这些
    1.41
    –
    1.37
    /*!
    1.37
    AFFIRMED
    1.32
    のでしょう
    1.31
     ordini
    1.30
    Activations Density 0.003%

    No Known Activations