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

    Neuronpedia

    INDEX
    Explanations

    phrases related to conflicts, disputes, and violence

    oai_token-act-pair · gpt-3.5-turbo

    ellipses and trailing punctuation indicating pauses or omissions in dialogue or narrative

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ord
    -0.77
    oresc
    -0.76
    odi
    -0.74
    flies
    -0.74
    inks
    -0.73
    icum
    -0.72
    urat
    -0.71
    anto
    -0.70
    iard
    -0.70
    ãĥīãĥ©
    -0.70
    POSITIVE LOGITS
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    0.89
    âĢ¦âĢ¦âĢ¦âĢ¦
    0.85
    ³³³³³³³³³³³³³³³³
    0.76
     Continue
    0.72
     Written
    0.69
     uh
    0.65
     ,"
    0.64
    ////////////////////////////////
    0.64
    .....
    0.63
     Wait
    0.61
    Act Density 0.040%

    No Known Activations