© 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. 10537
    Prev
    Next
    INDEX
    Explanations

    often, usually, sometimes, potentially

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

    The neuron is detecting adverbs and other degree or frequency modifiers that qualify or generalize statements.

    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
    0.95
    0.72
     \
    0.68
    、
    0.65
     The
    0.65
    The
    0.63
    0.63
    <0xE3>
    0.62
     other
    0.61
     American
    0.61
    POSITIVE LOGITS
     которой
    0.92
     което
    0.91
     которых
    0.89
     котором
    0.88
     включая
    0.87
    which
    0.86
     которого
    0.84
     которым
    0.84
     которому
    0.84
     jonka
    0.82
    Activations Density 0.355%

    No Known Activations