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

    Neuronpedia

    INDEX
    Explanations

    forum/blog posts

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on tokens ending in “ed,” i.e. past-tense or past-participle forms.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    맨
    -0.06
    [sub
    -0.06
     ArrayCollection
    -0.06
     Lion
    -0.06
     JAVA
    -0.06
    368
    -0.06
     Double
    -0.06
     Trường
    -0.06
     tendencies
    -0.06
     Hollande
    -0.06
    POSITIVE LOGITS
     sire
    0.07
     pace
    0.07
     Austral
    0.06
    190
    0.06
    (Size
    0.06
     hubby
    0.06
     pla
    0.06
    direct
    0.06
    Plot
    0.06
     Ow
    0.06
    Act Density 0.014%

    No Known Activations