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

    Neuronpedia

    INDEX
    Explanations

    prefer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Options
    0.49
     lemb
    0.49
     Senate
    0.47
     SAM
    0.47
     STR
    0.46
     Roundup
    0.45
     H
    0.45
     Graduate
    0.45
     LAM
    0.44
     WL
    0.44
    POSITIVE LOGITS
    terer
    0.57
    ti
    0.50
    使用
    0.49
    instead
    0.49
     방법을
    0.49
     അല്ലെങ്കിൽ
    0.48
    r
    0.48
    no
    0.48
    different
    0.48
     അല്ല
    0.48
    Act Density 0.004%

    No Known Activations