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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating various types of problems or issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    caloosa
    -0.38
     føl
    -0.37
     '{@
    -0.36
     canvas
    -0.35
     виправивши
    -0.31
    entrant
    -0.31
    wikimedia
    -0.30
     Hadid
    -0.30
     Vest
    -0.30
     Andi
    -0.30
    POSITIVE LOGITS
     problems
    1.17
     Problems
    1.17
    Problems
    1.09
    problems
    1.07
     troubles
    1.05
     Trouble
    1.04
    proble
    1.04
     problemen
    1.03
     PROBLEMS
    1.02
     Probleme
    1.02
    Act Density 0.425%

    No Known Activations