© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 16057
    Prev
    Next
    INDEX
    Explanations

    roles need to do something

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

    The main thing this neuron does is identify mentions of professional roles or named positions (e.g., directors, advisors, staff).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    announced
    1.00
     Announces
    0.99
    宣布
    0.99
    들도
    0.95
     объяви
    0.95
     ने
    0.93
    ’
    0.93
     anunció
    0.88
    ’
    0.88
     spokesman
    0.87
    POSITIVE LOGITS
     a
    1.13
     any
    1.05
     subjective
    1.03
     geometric
    1.00
     negative
    0.98
     the
    0.97
     an
    0.97
     unsaturated
    0.96
     diagonal
    0.95
     grayscale
    0.94
    Activations Density 0.000%

    No Known Activations