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

    Neuronpedia

    INDEX
    Explanations

    details related to news reports and police activities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    eatures
    -0.58
    awei
    -0.54
    olars
    -0.48
    ricanes
    -0.47
    ijing
    -0.46
    ixel
    -0.46
    eatured
    -0.46
    atars
    -0.45
    umbnails
    -0.44
    raits
    -0.44
    POSITIVE LOGITS
     Somebody
    0.58
     Especially
    0.54
     Alright
    0.54
     Yeah
    0.53
     Whatever
    0.53
     Totally
    0.52
     Unless
    0.52
     Someone
    0.52
     Lets
    0.51
     Something
    0.51
    Act Density 2.474%

    No Known Activations