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

    Neuronpedia

    INDEX
    Explanations

    sentences ending in punctuation marks

    oai_token-act-pair · gpt-3.5-turbo

    expressions of uncertainty or speculation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ÂŃ
    -0.78
     Mehran
    -0.72
    å§
    -0.66
    obil
    -0.62
    ensable
    -0.60
    worldly
    -0.58
    å½
    -0.58
    Thousands
    -0.57
     unrecogn
    -0.55
    ¨
    -0.55
    POSITIVE LOGITS
    Anyway
    1.19
     Anyway
    1.13
     anyways
    0.93
     tho
    0.90
     dunno
    0.90
    Still
    0.84
     kinda
    0.83
     honestly
    0.83
     definitely
    0.82
     aside
    0.77
    Act Density 1.042%

    No Known Activations