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

    Neuronpedia

    INDEX
    Explanations

    and

    np_max-act · gemini-2.0-flash

    This neuron responds to modal verbs and conditional markers (can, could, want, do, if) that signal asking about ability or possibility.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     силу
    -0.06
     depart
    -0.06
     ale
    -0.06
    ěř
    -0.06
     kuru
    -0.06
     NSMutableArray
    -0.06
     emb
    -0.06
     frail
    -0.06
     nejen
    -0.06
     (!$
    -0.06
    POSITIVE LOGITS
    /service
    0.09
    FormField
    0.07
     Survivor
    0.07
    _language
    0.07
     retains
    0.06
    Parameters
    0.06
     Confidence
    0.06
    _mini
    0.06
     workshop
    0.06
     monstrous
    0.06
    Act Density 0.117%

    No Known Activations