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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the word "on" and its variations in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    phrases starting with "on the" that introduce contrasting or additional perspectives.

    oai_token-act-pair · gpt-4oTriggered by @christianthto

    "on" followed by a determiner or "average"

    np_acts-logits-general · gemini-2.0-flash

    on average

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ManagedBean
    -0.66
    ModelSerializer
    -0.65
    jooq
    -0.59
     fondament
    -0.57
     iParam
    -0.56
    enumii
    -0.56
    ########.
    -0.54
    offsetof
    -0.54
    thansa
    -0.54
     uParam
    -0.53
    POSITIVE LOGITS
     behalf
    1.78
     top
    1.03
     occasion
    1.01
     the
    1.01
     account
    0.89
    omato
    0.87
     average
    0.82
     both
    0.81
     purpose
    0.80
     horseback
    0.78
    Act Density 0.377%

    No Known Activations