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

    Neuronpedia

    INDEX
    Explanations

    attends to the token "told" from various sources or quoted references

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.18
    1:0.21
    2:0.11
    3:0.06
    4:0.05
    5:0.03
    6:0.05
    7:0.27
    Negative Logits
     AssemblyTitle
    -0.50
     MonoBehaviour
    -0.45
    igshid
    -0.43
    verwijspagina
    -0.42
    fjspx
    -0.41
    +:+
    -0.41
    AddTagHelper
    -0.41
    rrggbb
    -0.41
    addGap
    -0.40
    addPreferredGap
    -0.40
    POSITIVE LOGITS
    sub
    0.35
    istic
    0.34
    изнь
    0.33
    ("//
    0.32
    　　“
    0.32
    erty
    0.32
    —“
    0.32
    lamation
    0.32
    Carla
    0.32
    izable
    0.31
    Act Density 0.026%

    No Known Activations