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

    Neuronpedia

    INDEX
    Explanations

    actions or commands related to controlling devices or software

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    This neuron detects the coordinating conjunction "and" (common connective/sequence marker in lists and clauses).

    oai_token-act-pair · gpt-5-miniTriggered by @phamvanluong2003
    New Auto-Interp
    Negative Logits
     AssemblyCompany
    -0.68
     насељу
    -0.65
    marginHorizontal
    -0.57
    flexGrow
    -0.55
    gemene
    -0.52
     AttributeSet
    -0.51
    CopyWith
    -0.50
    Scénario
    -0.50
    yarnpkg
    -0.49
     insuffisamment
    -0.49
    POSITIVE LOGITS
     Instantly
    1.06
     instantaneously
    0.91
     whereupon
    0.84
     instantly
    0.84
     thereupon
    0.79
     Commence
    0.76
     swarovski
    0.75
     Automatically
    0.74
     Incenti
    0.73
     wikihow
    0.73
    Act Density 0.401%

    No Known Activations