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

    Neuronpedia

    INDEX
    Explanations

    phrases related to personal characteristics and attributes, particularly in the context of identity and belonging

    oai_token-act-pair · gpt-3.5-turbo

    pronouns related to identity and group association

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    forcing
    -0.90
    forcement
    -0.86
     freeing
    -0.83
     concluding
    -0.76
     preventing
    -0.76
     temptation
    -0.73
     tightening
    -0.72
     retrieving
    -0.71
    limits
    -0.71
     forcing
    -0.70
    POSITIVE LOGITS
     belong
    1.47
     belonged
    1.39
     lived
    1.32
     reside
    1.25
     graduated
    1.22
     resided
    1.18
     specialize
    1.14
     hail
    1.13
     grew
    1.12
     married
    1.11
    Act Density 0.413%

    No Known Activations