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

    Neuronpedia

    INDEX
    Explanations

    Experimental results and comparisons

    np_max-act · gemini-2.0-flash

    mentions of experimental results, evaluations, or statements reporting that experiments/show/results demonstrate the method's performance.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     bene
    -0.07
     HttpSession
    -0.06
    onn
    -0.06
    -No
    -0.06
     арми
    -0.06
    kur
    -0.06
     identify
    -0.06
    yro
    -0.06
    ound
    -0.06
     können
    -0.06
    POSITIVE LOGITS
     провести
    0.07
    _TRNS
    0.06
     CDDL
    0.06
    _deinit
    0.06
     тщ
    0.06
    SHIP
    0.06
    \Model
    0.06
    劇
    0.06
    .logged
    0.06
    頼
    0.06
    Act Density 0.054%

    No Known Activations