© 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. 22-GEMMASCOPE-RES-131K
    7. 104401
    Prev
    Next
    INDEX
    Explanations

    events being held or performed

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

    This neuron fires on auxiliary verbs (especially “will,” “be,” and similar forms) used to announce or schedule events or actions.

    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
     нарушения
    -1.43
     who
    -1.26
     whose
    -1.23
    Such
    -1.22
     such
    -1.21
    不清楚
    -1.19
     their
    -1.17
     that
    -1.17
    Since
    -1.17
    其他
    -1.15
    POSITIVE LOGITS
     exacte
    1.31
     on
    1.30
     perilaku
    1.21
     bricola
    1.19
     is
    1.19
     alho
    1.19
     avocat
    1.18
    Scienti
    1.11
     perfecte
    1.10
    skie
    1.09
    Activations Density 0.001%

    No Known Activations