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

    Neuronpedia

    INDEX
    Explanations

    verbs in infinitive form

    oai_token-act-pair · gpt-3.5-turbo

    references to rotary actions or movements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    natureconservancy
    -0.86
    ãģŁ
    -0.69
    BILITIES
    -0.64
    ahime
    -0.62
     Occupations
    -0.61
    ¯¯
    -0.59
     quarters
    -0.59
     sudden
    -0.59
    ept
    -0.58
     CARE
    -0.58
    POSITIVE LOGITS
    rot
    1.08
    omy
    0.95
    unda
    0.93
    iful
    0.90
    ations
    0.89
    ica
    0.88
    ation
    0.87
    orus
    0.84
    ational
    0.83
    ate
    0.81
    Act Density 0.008%

    No Known Activations