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

    Neuronpedia

    INDEX
    Model
    gemma-2-9b-it
    Layer #
    20
    Steering Hook
    blocks.20.hook_resid_pre
    Steering Strength
    69
    Uploader
    bot-neuronpedia
    Created At
    2/15/2025 1:06:43 AM
    Raw Vector
    Actions
    Explanations

    instances of the word "said."

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Negative Logits
     Wiktionnaire
    -0.51
     Audiodateien
    -0.51
     &___
    -0.50
    chemise
    -0.49
     homoto
    -0.49
    HttpFoundation
    -0.49
    OGND
    -0.49
     surla
    -0.49
    WriteTagHelper
    -0.47
    bootstrapcdn
    -0.46
    POSITIVE LOGITS
     said
    0.58
     saying
    0.51
     told
    0.45
     řekl
    0.43
    said
    0.42
     say
    0.40
    ValueStyle
    0.40
     dijo
    0.39
     talking
    0.39
    ########.
    0.38
    Act Density 0.000%

    No Known Activations