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

    Neuronpedia

    INDEX
    Explanations

    strong and assertive phrases or commands

    oai_token-act-pair · gpt-3.5-turbo

    expressions of fear or anxiety

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     fragmented
    -0.84
     targeted
    -0.81
     fragmentation
    -0.76
     territorial
    -0.76
     synerg
    -0.75
     spir
    -0.75
     locality
    -0.75
     dubbed
    -0.75
     proport
    -0.75
     isolated
    -0.73
    POSITIVE LOGITS
    Yeah
    1.36
    Okay
    1.26
    Alright
    1.26
    Yes
    1.25
    Hey
    1.25
    yeah
    1.21
    Hmm
    1.18
    âĶĢâĶĢâĶĢâĶĢ
    1.16
    Huh
    1.16
    Oh
    1.15
    Act Density 0.146%

    No Known Activations