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

    Neuronpedia

    INDEX
    Explanations

    multi-word phrases that indicate structured or significant ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.01
    1:0.01
    2:0.06
    3:0.04
    4:0.07
    5:0.02
    6:0.17
    7:0.40
    8:0.03
    9:0.03
    10:0.06
    11:0.07
    Negative Logits
     typo
    -1.46
    yrics
    -1.45
    igmatic
    -1.45
    Fs
    -1.40
    aska
    -1.38
    imei
    -1.36
     sqor
    -1.34
    furt
    -1.33
    esan
    -1.27
    ensical
    -1.26
    POSITIVE LOGITS
     ceilings
    1.80
     horizont
    1.79
    clus
    1.66
     lineback
    1.58
    rity
    1.57
     antioxid
    1.56
     pillars
    1.54
     helic
    1.54
     horm
    1.49
     pillar
    1.44
    Act Density 0.001%

    No Known Activations