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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Dunefsky · Chlenski · Transcoders Enable Fine-Grained Interpretable Circuit Analysis
    3. GPT2-Small
    4. Transcoders Residuals
    5. 9-TRES-DC
    6. 653
    Prev
    Next
    INDEX
    Explanations

    components related to technology and design specifications

    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
    onite
    -0.73
     internationally
    -0.68
     Transparency
    -0.63
     Lauder
    -0.61
     Parables
    -0.58
    ioxide
    -0.57
     EVER
    -0.56
     twentieth
    -0.56
    ploma
    -0.55
     universes
    -0.55
    POSITIVE LOGITS
     permitting
    0.70
    Correct
    0.69
     providing
    0.68
     allowing
    0.66
     making
    0.64
     delivering
    0.64
    esc
    0.64
     and
    0.64
     disg
    0.64
     enabling
    0.63
    Activations Density 0.464%

    No Known Activations