© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66710
    Prev
    Next
    INDEX
    Explanations

    kind of

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

    The neuron fires on hedge or approximation phrases—e.g. “kind of,” “sort of,” “kind of like”—that soften or qualify statements.

    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
    焄
    -1.86
     NACIONAL
    -1.75
     Tatsache
    -1.70
     untuk
    -1.68
    when
    -1.63
    <
    -1.57
    瀦
    -1.56
     Notwendigkeit
    -1.55
    {
    -1.55
    zzlies
    -1.55
    POSITIVE LOGITS
     the
    1.59
    ’
    1.55
    feest
    1.41
     klokken
    1.40
    idlertid
    1.35
    Vox
    1.34
     
    1.33
    
    1.33
     knihy
    1.31
    >+</
    1.30
    Activations Density 0.006%

    No Known Activations