© 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. 72088
    Prev
    Next
    INDEX
    Explanations

    error messages or code context

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

    The neuron is strongly activated by tokens that mark code‐position or error‐trace numbers (e.g. line/column indices in stack traces or JSON code annotations).

    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
     Berufung
    -0.87
     prédé
    -0.87
     göm
    -0.84
    场上
    -0.82
     telefonu
    -0.79
    .))
    -0.78
     betalen
    -0.77
     ocultar
    -0.77
     coordinación
    -0.77
     apreciar
    -0.77
    POSITIVE LOGITS
    ordeel
    0.87
    おきます
    0.85
     immediately
    0.85
     スキン
    0.82
     entsprechende
    0.79
    спери
    0.77
    țial
    0.77
    ො
    0.77
    uterte
    0.77
    含量
    0.76
    Activations Density 0.001%

    No Known Activations