© 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. 22-GEMMASCOPE-RES-131K
    7. 38893
    Prev
    Next
    INDEX
    Explanations

    precedent by, then what next

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

    The neuron fires on modal and authority-related words—especially auxiliaries and verbs expressing ability, permission, obligation, or future action (e.g. will/would, can/should, authority, encourage).

    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
     директор
    -0.97
     hexagon
    -0.96
     rencontr
    -0.93
    layoutManager
    -0.90
    IMHO
    -0.90
    kington
    -0.89
    llion
    -0.88
     pasajeros
    -0.87
     rencontrer
    -0.87
    様に
    -0.87
    POSITIVE LOGITS
     every
    1.17
     one
    1.05
    ֻ
    1.02
     any
    0.99
    āci
    0.97
     just
    0.97
     first
    0.94
     incenti
    0.91
     limitless
    0.91
    Every
    0.90
    Activations Density 0.024%

    No Known Activations