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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. GPT2-Small
    3. Transcoders Residuals
    4. 8-TRES-DC
    5. 24
    Prev
    Next
    INDEX
    Explanations

    instances of the word "with" indicating varying degrees of anticipation or emotion

    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
    perse
    -0.78
    meet
    -0.78
    enne
    -0.69
    burgh
    -0.68
     hey
    -0.68
    wat
    -0.67
    Tap
    -0.65
    bush
    -0.63
    eas
    -0.63
    olit
    -0.63
    POSITIVE LOGITS
     emotion
    0.69
     speculation
    0.66
     grief
    0.63
     gratitude
    0.63
     angst
    0.62
     negativity
    0.61
     ambiguity
    0.61
     feedback
    0.60
     nostalgia
    0.59
     tension
    0.59
    Activations Density 0.107%

    No Known Activations