© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 38680
    Prev
    Next
    INDEX
    Explanations

    composition over inheritance

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

    This neuron detects comparative or preference expressions, especially “X over Y” phrasing (e.g. prefer, favor, composition over inheritance).

    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
    rigal
    -0.80
    żli
    -0.77
     alér
    -0.75
    ctuaries
    -0.75
     Für
    -0.75
    ErrorCode
    -0.74
     tiệc
    -0.73
    berte
    -0.72
    OnChange
    -0.71
    cillary
    -0.71
    POSITIVE LOGITS
     over
    4.75
     above
    3.41
     над
    2.78
     över
    2.36
    above
    2.33
    over
    2.19
     ahead
    2.05
     gegenüber
    2.03
     sobre
    2.02
     boven
    2.02
    Activations Density 0.079%

    No Known Activations