© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. GPT2-Small
    5. Transcoders Residuals
    7. 8-TRES-DC
    9. 821
    Prev
    Next
    INDEX
    Explanations

    questions and phrases directed at the audience, particularly the word "you."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Rica
    -0.70
    comes
    -0.67
    GGGGGGGG
    -0.63
     Fi
    -0.58
     flowed
    -0.57
    orned
    -0.57
    rified
    -0.57
    ivated
    -0.57
    rase
    -0.56
    orough
    -0.56
    POSITIVE LOGITS
     reckon
    0.78
     suppose
    0.75
     think
    0.73
     wanna
    0.73
    deals
    0.72
     recall
    0.70
     conceive
    0.70
     expect
    0.70
     practise
    0.69
     want
    0.68
    Activations Density 0.051%

    No Known Activations