© 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. 72464
    Prev
    Next
    INDEX
    Explanations

    programming keywords and symbols

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

    The neuron fires on CamelCase identifier tokens (e.g. class or component names) in code.

    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
     pelear
    -1.09
     și
    -0.97
     llorar
    -0.96
    歲
    -0.94
    bringer
    -0.92
    ÜN
    -0.91
     OGSÅ
    -0.91
    liothèque
    -0.91
    LocalTime
    -0.90
    bledon
    -0.90
    POSITIVE LOGITS
     Parole
    1.06
    かな
    0.92
     at
    0.86
     hatt
    0.85
     может
    0.82
    Nev
    0.81
     chaleur
    0.81
     of
    0.79
    コラボ
    0.78
    Sog
    0.78
    Activations Density 0.141%

    No Known Activations