© 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. 10-GEMMASCOPE-RES-131K
    7. 119581
    Prev
    Next
    INDEX
    Explanations

    fractional mathematical exponents

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

    The neuron never activates for any token—it appears to be effectively inert.

    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
    .
    -1.97
    to
    -1.84
    )
    -1.82
    %
    -1.56
    .”
    -1.55
     OGSÅ
    -1.51
    なんで
    -1.50
    煥
    -1.49
    other
    -1.49
    Ｐ
    -1.48
    POSITIVE LOGITS
    潆
    1.84
    ↵
    1.81
    聍
    1.68
    шают
    1.63
    
    1.62
    ǿ
    1.59
     fizik
    1.56
     nasce
    1.56
    我了
    1.55
    溷
    1.55
    Activations Density 0.024%

    No Known Activations