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

    Neuronpedia

    INDEX
    Explanations

    Mojo, Kivy, ALiBi

    np_acts-logits-general · gemini-2.5-flash-lite

    mentions of specific branded proper nouns, especially names of tech products, platforms, models, or companies.

    oai_token-act-pair · gpt-5Triggered by @yooniel31
    New Auto-Interp
    Negative Logits
    í
    0.38
     And
    0.34
    our
    0.33
    es
    0.33
     It
    0.33
    ure
    0.32
    ring
    0.31
    &
    0.31
    It
    0.30
    ret
    0.30
    POSITIVE LOGITS
    ﻜ
    0.35
    は
    0.31
    ﺛ
    0.30
     on
    0.29
    ﺿ
    0.29
     is
    0.29
     motores
    0.29
    ﺘ
    0.27
     perceive
    0.27
    ﭔ
    0.26
    Act Density 0.340%

    No Known Activations