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

    Neuronpedia

    INDEX
    Explanations

    elements related to cultural commentary and experiences

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

    Follows certain adjectives/adverbs

    np_acts-logits-general · gemini-2.0-flash

    truly has it all

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Helpful
    -0.67
     helpful
    -0.63
     interesting
    -0.59
    NewUrlParser
    -0.57
     useful
    -0.57
     Helpful
    -0.56
    helpful
    -0.55
    Useful
    -0.53
    Interesting
    -0.53
     Initially
    -0.52
    POSITIVE LOGITS
     couldn
    0.75
     truly
    0.74
    truly
    0.72
     virkelig
    0.70
     einfach
    0.70
    couldn
    0.67
     verkligen
    0.67
     ticks
    0.66
     parlent
    0.66
     définitivement
    0.65
    Act Density 0.148%

    No Known Activations