© 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. 34-GEMMASCOPE-RES-131K
    7. 92799
    Prev
    Next
    INDEX
    Explanations

    copyright and warranty information

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

    The neuron activates most strongly on tokens containing underscores—i.e. code identifiers or method names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     window
    -0.81
     second
    -0.75
     complicated
    -0.71
    niños
    -0.70
     працю
    -0.68
     driver
    -0.68
     what
    -0.67
     windows
    -0.66
    Děkuji
    -0.66
    getWindow
    -0.64
    POSITIVE LOGITS
    korb
    0.81
    ſhed
    0.79
     praxis
    0.78
     rowing
    0.77
     heath
    0.73
     لاع
    0.72
    bay
    0.71
    バッグ
    0.71
    ффек
    0.70
     californ
    0.70
    Activations Density 0.061%

    No Known Activations