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

    Neuronpedia

    INDEX
    Explanations

    phrases related to strong commitments or promises

    oai_token-act-pair · gpt-3.5-turbo

    instances of people or groups making commitments or promises

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     reproduction
    -0.64
     mix
    -0.60
     interaction
    -0.59
     curve
    -0.59
     abstract
    -0.58
     CC
    -0.58
     Mix
    -0.58
     feedback
    -0.57
     DK
    -0.57
     ping
    -0.57
    POSITIVE LOGITS
     vowed
    3.14
     vows
    2.10
     swore
    1.77
     pledged
    1.72
     vow
    1.65
     promised
    1.53
     insisted
    1.32
     thanked
    1.30
     prayed
    1.27
     apologized
    1.26
    Act Density 0.032%

    No Known Activations