Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    instances of personal experience and technology-related discussions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    past participles and following nouns

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

    The neuron activates on verbs describing actions (especially past-tense and present-participial forms like "developing," "studied," "taking," "started," "ran/ran").

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    
    -0.40
    istisches
    -0.38
    uxxxx
    -0.35
     VizieR
    -0.35
    WriteTagHelper
    -0.33
     הרי
    -0.33
     '\\;'
    -0.33
    허
    -0.32
    Menurutnya
    -0.32
    ÍST
    -0.32
    POSITIVE LOGITS
    SourceChecksum
    0.50
     $_(
    0.48
     препратки
    0.46
     Geſch
    0.45
    endwhile
    0.44
     Conclu
    0.44
    PostInfinity
    0.44
    pulos
    0.43
     ControllerBase
    0.43
     ſche
    0.43
    Act Density 0.175%

    No Known Activations