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

    Neuronpedia

    INDEX
    Explanations

    verbs related to actions or processes

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to processes or actions that require involvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ammunition
    -0.64
    mith
    -0.62
     Keys
    -0.62
    grim
    -0.62
     kindly
    -0.59
     Ninth
    -0.59
    ye
    -0.58
    Ward
    -0.57
     Editors
    -0.56
    gh
    -0.55
    POSITIVE LOGITS
     involves
    3.07
     involve
    2.90
     entails
    2.13
     involving
    1.87
     entail
    1.80
     involved
    1.72
     revolves
    1.58
     consists
    1.50
    involved
    1.38
     consisted
    1.35
    Act Density 0.019%

    No Known Activations