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

    Neuronpedia

    INDEX
    Explanations

    code comments and references to error handling and file operations

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

    Follow-up actions like "continue", "skip", or "ignore"

    np_acts-logits-general · gemini-2.0-flash

    stopping, ignoring, or rejecting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    essoal
    -0.53
    意外と
    -0.45
    GENT
    -0.43
    itai
    -0.43
    chelles
    -0.42
    Perd
    -0.41
     rubbing
    -0.40
    Cada
    -0.40
    roto
    -0.40
     interrelated
    -0.40
    POSITIVE LOGITS
     continue
    0.84
     skip
    0.78
    continue
    0.76
     nothing
    0.73
    Portale
    0.73
     considérons
    0.73
     abort
    0.72
     Continue
    0.72
    AddTagHelper
    0.72
     ignore
    0.71
    Act Density 0.496%

    No Known Activations