© 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. 68647
    Prev
    Next
    INDEX
    Explanations

    business, bill, q, /*

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

    This neuron detects formatting or markup control tokens—such as backslashes, escape sequences, and numeric‐code parameters—in structured document or code headers.

    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
     their
    -1.94
     both
    -1.91
     not
    -1.68
     even
    -1.67
     at
    -1.67
     or
    -1.64
     people
    -1.57
     blatantly
    -1.56
     from
    -1.50
     subtly
    -1.48
    POSITIVE LOGITS
     ส์
    1.36
    could
    1.35
     trám
    1.34
    primarily
    1.34
    commonly
    1.32
     geprüft
    1.31
     alcuni
    1.28
     パッド
    1.27
     inscripción
    1.27
    küche
    1.26
    Activations Density 0.000%

    No Known Activations