© 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. 10-GEMMASCOPE-RES-131K
    7. 59093
    Prev
    Next
    INDEX
    Explanations

    code setup actions

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

    The neuron is effectively never activated by any of these code‐snippet tokens—it doesn’t pick out any particular pattern and remains off (all zeros).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     involucrados
    -2.06
    \
    -2.03
     destacados
    -1.97
     efectivos
    -1.96
     listos
    -1.95
     afirmar
    -1.86
     aseguran
    -1.80
     evolucion
    -1.79
     dispuestos
    -1.79
    '
    -1.74
    POSITIVE LOGITS
    錸
    2.27
     sula
    2.17
    pography
    2.09
    遄
    2.00
     maris
    1.98
     laurent
    1.95
     gaun
    1.94
     siena
    1.94
     petra
    1.91
    "_
    1.91
    Activations Density 0.057%

    No Known Activations