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

    Neuronpedia

    INDEX
    Explanations

    phrases related to replacement or substitution

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.05
    2:0.26
    3:0.05
    4:0.03
    5:0.04
    6:0.12
    7:0.03
    8:0.03
    9:0.19
    10:0.03
    11:0.02
    Negative Logits
     Kafka
    -3.27
     chill
    -3.24
    amy
    -3.17
     clouds
    -3.14
     Guinness
    -3.11
     zoning
    -3.05
     Blaz
    -2.96
     Booth
    -2.95
     voy
    -2.91
     secrecy
    -2.89
    POSITIVE LOGITS
     replacements
    8.26
     Replacement
    7.96
     replacement
    7.85
    Repl
    7.54
    repl
    7.10
     Replace
    7.04
     replace
    6.84
     replaced
    6.70
     Repl
    6.54
    replace
    6.41
    Act Density 0.023%

    No Known Activations