© 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. 11280
    Prev
    Next
    INDEX
    Explanations

    as you, as is, as one, as reflected, as some, as long, as of, as it, as if

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

    This neuron selectively detects the subordinating conjunction “as” when it introduces comparative or explanatory clauses (e.g. “As far as I can see,”).

    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
    '
    0.77
    ',
    0.71
    ing
    0.71
    tter
    0.70
    এসসি
    0.68
    s
    0.68
    ING
    0.67
    ’,
    0.67
    '],
    0.66
    %,
    0.65
    POSITIVE LOGITS
    たとえば
    0.81
     Leibn
    0.79
     činjen
    0.77
     peuple
    0.75
     örne
    0.74
    bbene
    0.72
    cribing
    0.71
     Oxygen
    0.71
    麽
    0.71
    sembles
    0.71
    Activations Density 0.063%

    No Known Activations