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

    Neuronpedia

    INDEX
    Explanations

    references to the term "Ha." This could possibly be related to the use of the prefix in names, titles, or abbreviations

    oai_token-act-pair · gpt-3.5-turbo

    references to the Haqqani network

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    */(
    -0.90
    eers
    -0.83
    itives
    -0.80
    eenth
    -0.79
    hips
    -0.79
    ments
    -0.76
    mented
    -0.75
    âķIJ
    -0.73
    icipated
    -0.73
    rations
    -0.72
    POSITIVE LOGITS
     Ha
    1.12
    Ha
    1.12
    HAHA
    0.98
    iku
    0.92
    ifa
    0.85
    irst
    0.84
    ibi
    0.80
    verty
    0.79
    vel
    0.77
    HAHAHAHA
    0.77
    Act Density 0.005%

    No Known Activations