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

    Neuronpedia

    INDEX
    Explanations

    negative evaluations or assessments of situations or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    contrasting words however

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (°
    -0.42
    CreateModel
    -0.38
    出版年
    -0.34
    ({
    -0.32
    modb
    -0.32
     ({
    -0.31
     disso
    -0.31
    OptionsMenu
    -0.30
     ExecuteAsync
    -0.30
     Reverso
    -0.30
    POSITIVE LOGITS
     however
    0.94
     però
    0.91
    however
    0.88
     entanto
    0.84
     porém
    0.76
     cependant
    0.75
    却
    0.74
     jednak
    0.74
     jedoch
    0.70
     kuitenkin
    0.70
    Act Density 0.517%

    No Known Activations