© 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. 129075
    Prev
    Next
    INDEX
    Explanations

    facing opponents

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

    This neuron detects language about competitive matchups—mentions of opponents, chances or opportunities in a sports preview or tournament context.

    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
     aban
    -1.35
     coser
    -1.34
    beutel
    -1.33
     Tarjeta
    -1.33
     transporta
    -1.30
     monast
    -1.30
     decoracion
    -1.29
     cardin
    -1.29
    yı
    -1.27
     sembl
    -1.27
    POSITIVE LOGITS
     for
    1.87
     we
    1.77
     each
    1.49
     in
    1.45
     any
    1.38
     our
    1.36
     to
    1.32
    ing
    1.31
            
    1.30
     particular
    1.26
    Activations Density 0.043%

    No Known Activations