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

    Neuronpedia

    INDEX
    Explanations

    tender

    np_max-act · gemini-2.0-flash

    The neuron activates on words that describe cooking outcomes or food textures (e.g. golden, tender, melted, thickened, bubbly, mushy).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     долго
    -0.07
    tl
    -0.07
    างว
    -0.06
    Textures
    -0.06
     hours
    -0.06
     textarea
    -0.06
    ременно
    -0.06
    apl
    -0.06
    ================================================
    -0.06
     एव
    -0.06
    POSITIVE LOGITS
     trust
    0.07
     disruption
    0.07
    -redux
    0.06
    тє
    0.06
    Brad
    0.06
    (NO
    0.06
     Rubio
    0.06
     algumas
    0.06
     disrupted
    0.06
     sco
    0.06
    Act Density 0.006%

    No Known Activations