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

    Neuronpedia

    INDEX
    Explanations

    brackets and braces

    np_max-act · gemini-2.0-flash

    This neuron detects template/data-binding expressions enclosed in braces (e.g. `{item.name}`, `{{ item.image }}`, Svelte `{#each …}`/`{#if …}` blocks).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    arest
    -0.07
     automobile
    -0.07
     Foundations
    -0.07
     घटन
    -0.06
     ud
    -0.06
     Heads
    -0.06
     getContentPane
    -0.06
     Ingen
    -0.06
     acceptance
    -0.06
     Super
    -0.06
    POSITIVE LOGITS
    \base
    0.07
    leur
    0.06
    telegram
    0.06
    _cf
    0.06
    >#
    0.06
    >{{$
    0.06
    ">{{$
    0.06
    レイ
    0.06
     skype
    0.06
    .prof
    0.06
    Act Density 0.008%

    No Known Activations