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

    Neuronpedia

    INDEX
    Explanations

    words and phrases that express peculiarity or curiosity

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

    something is unusual

    np_acts-logits-general · gemini-2.0-flash

    odd, strange, interesting, weird

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     étoit
    -0.63
    شرين
    -0.62
     tertinggi
    -0.61
     étoient
    -0.56
     operativa
    -0.54
     auroit
    -0.53
    lossians
    -0.53
     assistenza
    -0.52
    WritableDatabase
    -0.52
     genieten
    -0.52
    POSITIVE LOGITS
    strange
    0.98
     strange
    0.98
     wierd
    0.90
     étrange
    0.89
    interesting
    0.89
    Interesting
    0.89
     weird
    0.89
     odd
    0.88
     Strange
    0.86
     quirks
    0.84
    Act Density 0.143%

    No Known Activations