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

    Neuronpedia

    INDEX
    Explanations

    safety and caution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     almonds
    0.59
     muscles
    0.56
     grinding
    0.54
     jScrollPane
    0.54
     prostitutes
    0.54
     kangaroos
    0.53
     crushed
    0.52
     destroys
    0.52
     proiz
    0.52
     acorns
    0.52
    POSITIVE LOGITS
    safety
    0.89
    安全性
    0.89
     সতর্কতা
    0.89
    Safety
    0.87
    的安全
    0.86
     precau
    0.86
     cautioned
    0.83
     జాగ్ర
    0.82
    afety
    0.81
    安全
    0.80
    Act Density 0.001%

    No Known Activations