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

    Neuronpedia

    INDEX
    Explanations

    completion, finishing, or completed

    np_max-act · gemini-2.0-flash

    narratives related to personal growth and career transitions.

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

    The neuron detects mentions of finishing or completing activities, i.e. verbs and phrases that indicate wrapping up or concluding tasks.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    <Base
    -0.07
     boon
    -0.07
    일
    -0.07
     Relationship
    -0.07
    Intent
    -0.06
    _store
    -0.06
     ambassador
    -0.06
     pores
    -0.06
     Brazilian
    -0.06
     aider
    -0.06
    POSITIVE LOGITS
     completed
    0.12
     completing
    0.08
    completed
    0.07
     vybav
    0.07
     обязатель
    0.06
    修
    0.06
    ,cv
    0.06
    ฆ
    0.06
    .visitInsn
    0.06
     블
    0.06
    Act Density 0.033%

    No Known Activations