© 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. 63930
    Prev
    Next
    INDEX
    Explanations

    starts of diverse data segments

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

    The neuron detects JSON or similar metadata field names (object properties) in structured data.

    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
     at
    -1.02
     emphasizes
    -0.93
     before
    -0.90
     profoundly
    -0.85
     sebagainya
    -0.85
    핸
    -0.84
     emphasizing
    -0.84
     or
    -0.83
     both
    -0.82
     epitom
    -0.82
    POSITIVE LOGITS
     Ähnliche
    1.09
     SPECIFIC
    0.96
     Häufig
    0.95
     QUICK
    0.92
    ’
    0.92
     Dennoch
    0.90
    🍐
    0.90
     häufigsten
    0.88
     heiligen
    0.88
     eigentlichen
    0.87
    Activations Density 0.000%

    No Known Activations