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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Llama3.3-70B-IT
    5. 50-RESID-POST-GF
    7. 54479
    Prev
    Next
    INDEX
    Explanations

    protection and blocking

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.3-70B-IT @ 50-resid-post-gf
    Configuration
    Goodfire/Llama-3.3-70B-Instruct-SAE-l50/Llama-3.3-70B-Instruct-SAE-l50.pt
    Prompts (Dashboard)
    10,000 prompts, 128 tokens each
    Dataset (Dashboard)
    lmsys/lmsys-chat-1m
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     loos
    -0.10
     avoidance
    -0.09
     insured
    -0.09
     caut
    -0.09
    æī¶
    -0.09
    ilan
    -0.09
     Kurd
    -0.09
    erne
    -0.09
     faç
    -0.09
     accommod
    -0.09
    POSITIVE LOGITS
     protection
    0.39
     protect
    0.33
     Protection
    0.32
    Protection
    0.30
    ä¿ĿæĬ¤
    0.29
     protecting
    0.29
     protects
    0.29
    protect
    0.27
     Protect
    0.26
    ä¿ĿèŃ·
    0.26
    Activations Density 0.092%

    No Known Activations