© 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. 33159
    Prev
    Next
    INDEX
    Explanations

    code structure or syntax

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

    The neuron flags tokens associated with error or failure outcomes—things like “not graphical,” “false,” “ERROR,” “Fatal,” “quit,” or other abort/exception indicators.

    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
     to
    -1.16
     you
    -1.15
     meren
    -1.02
     meur
    -1.00
     kolek
    -1.00
    Recv
    -0.98
     klient
    -0.97
     autonome
    -0.96
     televiz
    -0.96
     curieux
    -0.95
    POSITIVE LOGITS
     then
    0.93
    Теперь
    0.91
    enumi
    0.90
     trasporto
    0.90
    ");
    0.87
    每
    0.85
     further
    0.84
    .
    0.84
     انا
    0.84
    Morfologia
    0.83
    Activations Density 0.001%

    No Known Activations