© 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. 99992
    Prev
    Next
    INDEX
    Explanations

    been followed by context

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

    The neuron activates on verbs (especially auxiliary “be” or “have” forms and their participles/gerunds) that signal ongoing or continuous states or actions (e.g. “has been,” “specialising,” “lived,” “cooking”).

    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
     new
    -1.97
     because
    -1.57
    现在的
    -1.43
     meticulous
    -1.42
     newly
    -1.42
     Sometimes
    -1.39
     slightly
    -1.35
     newer
    -1.34
     those
    -1.34
     burgeoning
    -1.34
    POSITIVE LOGITS
    since
    1.62
     salade
    1.48
    ciate
    1.48
    糌
    1.38
    鹌
    1.34
    ⩕
    1.33
    自体は
    1.30
     administração
    1.28
     zarówno
    1.28
     incluso
    1.28
    Activations Density 0.158%

    No Known Activations