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

    Neuronpedia

    INDEX
    Explanations

    trainer and training

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    thrift
    0.47
     vyb
    0.44
    esian
    0.43
     sensitized
    0.43
    渴
    0.43
    敛
    0.41
     thirsty
    0.41
    iciency
    0.39
     decayed
    0.38
     Loew
    0.38
    POSITIVE LOGITS
     tokenizer
    0.55
    Tokenizer
    0.48
     Trainer
    0.45
     token
    0.42
     trainer
    0.41
    ArgumentParser
    0.41
    Trainer
    0.41
    热
    0.41
    trainer
    0.40
    newToken
    0.40
    Act Density 0.004%

    No Known Activations