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

    Neuronpedia

    INDEX
    Explanations

    references to objects made of glass

    oai_token-act-pair · gpt-3.5-turbo

    references to glass objects and materials

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Akin
    -0.76
     Ake
    -0.74
     Hussein
    -0.71
    CVE
    -0.70
     NCT
    -0.64
    ghan
    -0.64
     Statistical
    -0.63
     Tup
    -0.62
     Hussain
    -0.60
    IRED
    -0.60
    POSITIVE LOGITS
    door
    1.08
    ware
    0.99
     shards
    0.96
    glass
    0.95
     beads
    0.94
     bottles
    0.94
    works
    0.94
     jars
    0.93
    gow
    0.92
    bl
    0.91
    Act Density 0.018%

    No Known Activations