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

    Neuronpedia

    INDEX
    Explanations

    intensifiers or modifiers that express degrees of magnitude or comparison

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

    that intensify comparisons

    np_acts-logits-general · gemini-2.0-flash

    much and bit comparisons

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Jefus
    -0.94
     Theſe
    -0.87
     Chriftian
    -0.83
     becauſe
    -0.82
     Majefty
    -0.81
     ftate
    -0.79
     ſeveral
    -0.78
     ſtate
    -0.77
     myſelf
    -0.75
     Eſ
    -0.74
    POSITIVE LOGITS
     able
    0.80
     worth
    0.76
     likely
    0.72
     ready
    0.72
     seen
    0.70
     taken
    0.62
     capable
    0.62
     about
    0.61
     evident
    0.59
     like
    0.58
    Act Density 0.293%

    No Known Activations