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

    Neuronpedia

    INDEX
    Explanations

    geographical direction terms related to locations

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

    cardinal directions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     queſta
    -1.19
    <unused41>
    -1.12
    <unused8>
    -1.12
    <pad>
    -1.12
    <unused43>
    -1.12
    <unused79>
    -1.12
    <unused14>
    -1.11
    <unused16>
    -1.11
    <unused3>
    -1.11
    [@BOS@]
    -1.11
    POSITIVE LOGITS
    ↵↵
    0.34
     and
    0.28
    <eos>
    0.26
     
    0.25
     umum
    0.25
     inmediato
    0.25
     triunfo
    0.24
    ↵
    0.24
     terrenos
    0.24
      
    0.23
    Act Density 0.010%

    No Known Activations