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

    Neuronpedia

    INDEX
    Explanations

    references to power dynamics and control within narratives

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

    references to specific sports teams or player performance

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

    steal data

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     moest
    -0.57
     gyhoeddwyd
    -0.56
     struggling
    -0.56
     lost
    -0.55
     struggled
    -0.53
     fallen
    -0.52
    lost
    -0.51
     moeten
    -0.51
     menghadapi
    -0.50
     losing
    -0.50
    POSITIVE LOGITS
     steal
    0.69
     steals
    0.69
     réuss
    0.64
     stole
    0.64
    OrBuilder
    0.61
     stealing
    0.60
     lợi
    0.59
     gaining
    0.58
     hijack
    0.58
    EndTag
    0.58
    Act Density 0.312%

    No Known Activations