© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73136
    Prev
    Next
    INDEX
    Explanations

    programming errors and structure

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

    The neuron activates on programming identifiers—especially class/type names, constants, and other CamelCase or ALL-CAP tokens.

    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
    🤔
    -1.09
     espet
    -1.09
     preuß
    -1.08
    ípio
    -1.00
    setzung
    -0.99
     Verbesserung
    -0.98
     Erfüllung
    -0.96
    zsef
    -0.95
    faccia
    -0.95
     latter
    -0.94
    POSITIVE LOGITS
     maîtres
    1.05
     your
    1.02
     weißem
    1.00
    报记者
    1.00
    他要
    0.98
     aparato
    0.96
    着的
    0.94
    ötet
    0.93
    ਫ਼
    0.92
    imágenes
    0.92
    Activations Density 0.089%

    No Known Activations