© 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. 22-GEMMASCOPE-RES-131K
    7. 16161
    Prev
    Next
    INDEX
    Explanations

    modal verbs followed by states or occurrences

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

    The neuron specializes in spotting modal or auxiliary verbs that express necessity or possibility (e.g. needs, must, can, will).

    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
    衬衫
    -1.24
    hamed
    -1.09
     didn
    -1.08
    eapples
    -1.05
    cended
    -1.05
    kwij
    -1.05
     بف
    -1.05
    eien
    -1.05
    which
    -1.04
    hibe
    -1.04
    POSITIVE LOGITS
     there
    1.31
     occur
    1.27
     '\\;'
    1.24
     arşivlendi
    1.15
    vailable
    1.15
     occurred
    1.15
     hadir
    1.13
     exists
    1.13
     occurs
    1.13
     héroe
    1.10
    Activations Density 0.071%

    No Known Activations