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

    Neuronpedia

    INDEX
    Explanations

    specific numerical values and their associations within contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Tokens after certain mathematical/scientific terms

    np_acts-logits-general · gemini-2.0-flash

    conjunctions and adverbs indicating concession

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    სქოლიო
    -0.38
    Au
    -0.35
    Resource
    -0.34
    Duty
    -0.34
    Archi
    -0.34
     AssemblyCulture
    -0.33
    OPTIM
    -0.32
     opti
    -0.32
    AnchorTagHelper
    -0.32
    Replication
    -0.32
    POSITIVE LOGITS
    ContentAsync
    0.70
     nevertheless
    0.59
     anyway
    0.56
     nonetheless
    0.56
     Administrativna
    0.56
     anyhow
    0.55
    それでも
    0.55
     anyways
    0.54
    Hentet
    0.53
    endphp
    0.52
    Act Density 1.319%

    No Known Activations