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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. GPT2-Small
    3. Transcoders Residuals
    4. 8-TRES-DC
    5. 951
    Prev
    Next
    INDEX
    Explanations

    phrases indicating actions or measures being taken

    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
    Login
    -0.68
     Rhod
    -0.66
    agram
    -0.62
     Elite
    -0.61
     fraudulent
    -0.60
     CGI
    -0.60
     motion
    -0.57
    WER
    -0.56
     destructive
    -0.56
     Rochester
    -0.56
    POSITIVE LOGITS
     preserve
    0.84
    bring
    0.84
     assure
    0.83
    appropriately
    0.83
     insure
    0.82
     ensure
    0.81
     discourage
    0.80
     reassure
    0.80
     liberate
    0.79
    veyard
    0.79
    Activations Density 0.154%

    No Known Activations