© 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. 31-GEMMASCOPE-2-RES-16K
    7. 454
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to the indefinite article “a” when it appears in a copular definition or description (e.g. “X is a …,” “I am not a …,” “It is a …”).

    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_31_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
     czyli
    0.33
     (=
    0.31
     adlı
    0.30
     ialah
    0.30
     (†
    0.30
     MPa
    0.30
     alebo
    0.29
    \,.
    0.29
     〢
    0.29
     Koordin
    0.28
    POSITIVE LOGITS
     veritable
    0.54
     throwback
    0.45
     boon
    0.43
     quintessential
    0.43
     prelude
    0.43
     rarity
    0.43
     reminder
    0.42
     deterrent
    0.42
     testament
    0.41
     catalyst
    0.41
    Activations Density 1.196%

    No Known Activations