© 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. 21583
    Prev
    Next
    INDEX
    Explanations

    exceptional positive attributes

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

    The neuron strongly activates on the word “exceptional,” i.e. it detects that specific high-praise adjective.

    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
    
    -3.28
    嘖
    -3.20
    
    -3.17
    嫑
    -3.08
    猡
    -2.92
    嫫
    -2.89
    僦
    -2.84
    腠
    -2.84
    蠖
    -2.84
    mpagne
    -2.84
    POSITIVE LOGITS
    There
    3.94
    While
    3.83
    Although
    3.77
    Even
    3.61
    (?)
    3.61
    Now
    3.59
    With
    3.58
    Some
    3.45
    Many
    3.41
    However
    3.41
    Activations Density 0.003%

    No Known Activations