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

    Neuronpedia

    INDEX
    Explanations

    phrases relating to deterioration and improvement in quality or performance

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

    worse or better comparison

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    SequentialGroup
    -0.44
     dimethyl
    -0.43
     legal
    -0.42
     redundant
    -0.41
    departure
    -0.41
     curie
    -0.40
    vertisers
    -0.40
     techn
    -0.40
    }]
    -0.40
     casualty
    -0.39
    POSITIVE LOGITS
     schlechter
    0.84
     schlechte
    0.82
     dår
    0.79
     dårlig
    0.76
    Worse
    0.75
     peores
    0.73
     schlecht
    0.72
    worse
    0.71
    Bad
    0.70
     Worse
    0.69
    Act Density 0.219%

    No Known Activations