© 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. 73157
    Prev
    Next
    INDEX
    Explanations

    comparison and superiority

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

    The neuron fires on comparative and superlative product-marketing language—phrases and words that position one item as “first,” “largest,” “better than,” “no comparison,” “offers,” etc., to highlight superiority.

    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
     could
    -1.08
     Sometimes
    -1.05
     occasionally
    -1.05
    民眾
    -1.04
     trochu
    -1.02
     but
    -0.94
     manque
    -0.94
     primarily
    -0.93
     mauvais
    -0.93
     seemed
    -0.91
    POSITIVE LOGITS
     competitors
    1.39
    competi
    1.16
     philippe
    1.15
    
    1.14
     compared
    1.13
     competitor
    1.13
     competi
    1.12
     dibanding
    1.07
     comparable
    1.05
     entidade
    1.05
    Activations Density 0.070%

    No Known Activations