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

    Neuronpedia

    INDEX
    Explanations

    modal verbs reflecting necessity or permission

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    giro
    -0.52
     tomorrow
    -0.48
     capacity
    -0.45
    >//
    -0.45
    Rhestr
    -0.43
    piracy
    -0.43
     unqualified
    -0.43
     Antiquity
    -0.42
    oon
    -0.42
     medically
    -0.42
    POSITIVE LOGITS
     war
    0.80
     hatte
    0.78
     waren
    0.78
     kam
    0.77
     ging
    0.77
     gab
    0.75
     konnte
    0.74
     gingen
    0.74
     kamen
    0.74
     hatten
    0.73
    Act Density 0.046%

    No Known Activations