© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130723
    Prev
    Next
    INDEX
    Explanations

    modal verbs

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

    The neuron fires on auxiliary/modal verbs (e.g. should, could, can, would, may, need, will), marking expressions of possibility, necessity, or permission.

    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
     [&
    -1.39
     Brussel
    -1.38
     legendar
    -1.38
    tta
    -1.38
     topp
    -1.35
     Kindergarten
    -1.32
    dyž
    -1.32
     Montessori
    -1.31
     monstre
    -1.30
     flotte
    -1.27
    POSITIVE LOGITS
     from
    1.81
     the
    1.52
     your
    1.38
     on
    1.34
     his
    1.33
     each
    1.27
     it
    1.27
     that
    1.25
     arbeitete
    1.23
     our
    1.22
    Activations Density 0.077%

    No Known Activations