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

    Neuronpedia

    INDEX
    Explanations

    baseball bases and positions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     underlies
    0.41
     Rot
    0.39
     rot
    0.39
     revolved
    0.38
     ე
    0.38
    ื่อ
    0.37
    reat
    0.37
    rots
    0.37
     তির
    0.36
    rot
    0.36
    POSITIVE LOGITS
     sack
    0.64
    垒
    0.63
     bag
    0.62
     bags
    0.59
    袋
    0.59
    壘
    0.59
    bag
    0.54
    bags
    0.54
     sacks
    0.51
     ব্যাগ
    0.51
    Act Density 0.001%

    No Known Activations