© 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. 14196
    Prev
    Next
    INDEX
    Explanations

    crumbling stone

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

    This neuron fires on common function words (especially articles and prepositions like “to,” “the,” “of,” “an,” “certain,” etc.) that introduce new clauses or phrases.

    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
    穵
    0.96
     rabbis
    0.95
     Супер
    0.93
     divertido
    0.90
     bốn
    0.88
     жі
    0.86
     divertida
    0.86
    ്ലാ
    0.83
     agregar
    0.83
     ricevuto
    0.83
    POSITIVE LOGITS
    if
    0.92
    ↵
    0.86
    [
    0.84
    ic
    0.81
    of
    0.80
    w
    0.77
    on
    0.77
    ut
    0.75
    c
    0.75
    it
    0.74
    Activations Density 0.000%

    No Known Activations