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

    Neuronpedia

    INDEX
    Explanations

    past tense "be" verbs

    np_max-act · gemini-2.0-flash

    narratives about personal experiences and reflections related to sports and coaching.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on “had” + verb constructions (i.e. past‐perfect verb phrases).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     aforementioned
    -0.07
     Hernandez
    -0.07
     blush
    -0.07
    Backup
    -0.06
    ुग
    -0.06
    มาก
    -0.06
    (up
    -0.06
    єї
    -0.06
     llama
    -0.06
    Russian
    -0.06
    POSITIVE LOGITS
     ناحیه
    0.07
     lodash
    0.06
    _log
    0.06
    _mE
    0.06
    .',
    0.06
    ="@
    0.06
    하우
    0.06
     　　　　　　
    0.06
    phys
    0.06
    DROP
    0.06
    Act Density 0.033%

    No Known Activations