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

    Neuronpedia

    INDEX
    Explanations

    think

    np_max-act · gemini-2.0-flash

    The neuron detects first‐person opinion or hedging phrases (e.g. “I think,” “I feel,” “I’m afraid,” etc.) that introduce subjective commentary.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    离
    -0.07
    以后
    -0.07
    addon
    -0.07
    720
    -0.07
     notas
    -0.07
    沟
    -0.07
    55
    -0.07
    seller
    -0.06
    */),
    -0.06
     مثل
    -0.06
    POSITIVE LOGITS
    	  ↵
    0.08
     Seems
    0.08
     think
    0.08
    [channel
    0.07
    ↵				↵
    0.07
    [c
    0.07
      ↵
    0.07
     시행
    0.07
     burns
    0.07
     Rahmen
    0.07
    Act Density 0.033%

    No Known Activations