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

    Neuronpedia

    INDEX
    Explanations

    words related to legal proceedings and medical conditions with negative connotations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Britann
    -0.55
    ICLE
    -0.55
    ĸļ
    -0.53
     anew
    -0.52
    ynthesis
    -0.50
     FACE
    -0.50
    orsche
    -0.50
    sha
    -0.48
    tex
    -0.46
    ynt
    -0.46
    POSITIVE LOGITS
    ables
    0.59
    osponsors
    0.56
    abled
    0.56
     absentee
    0.52
    åŃIJ
    0.50
    able
    0.50
     penalties
    0.49
    ability
    0.49
    ptives
    0.49
    Exception
    0.49
    Act Density 7.083%

    No Known Activations