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

    Neuronpedia

    INDEX
    Explanations

    command line history

    np_max-act · gemini-2.0-flash

    The neuron activates for tokens referring to remembering or recalling past conversation context or memory.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     resonance
    -0.07
    _cols
    -0.07
    agedList
    -0.07
    _recipe
    -0.06
     |:
    -0.06
     Jaguars
    -0.06
    /static
    -0.06
     solidity
    -0.06
     encaps
    -0.06
     endurance
    -0.06
    POSITIVE LOGITS
    信息
    0.07
    .isValid
    0.06
     чер
    0.06
    COM
    0.06
    Slave
    0.06
    Truth
    0.06
    être
    0.06
    gnore
    0.06
     Verify
    0.06
     notions
    0.06
    Act Density 0.033%

    No Known Activations