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

    Neuronpedia

    INDEX
    Explanations

    sections with special characters or coding syntax like underscores and double underscores

    oai_token-act-pair · gpt-3.5-turbo

    programming-related terminology or structures, particularly those associated with podcasts or initialization functions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Wooden
    -0.80
     Dynamics
    -0.68
     Noir
    -0.67
    osal
    -0.67
    oms
    -0.66
     Ravens
    -0.65
     Mour
    -0.64
    erie
    -0.64
     Instr
    -0.64
     Mate
    -0.63
    POSITIVE LOGITS
    dict
    0.91
    _.
    0.86
    ELL
    0.84
    future
    0.81
    DATA
    0.79
    oulder
    0.78
    vu
    0.78
    _-
    0.78
    ãĥĦ
    0.78
    init
    0.77
    Act Density 0.027%

    No Known Activations