© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Llama3.3-70B-IT
    5. 50-RESID-POST-GF
    7. 43529
    Prev
    Next
    INDEX
    Explanations

    protect and respect

    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
    oke
    -0.11
    arton
    -0.09
     Accept
    -0.09
    Insn
    -0.09
    ï¾ĺ
    -0.08
    alem
    -0.08
     numel
    -0.08
     accepting
    -0.08
     tavs
    -0.08
     democr
    -0.08
    POSITIVE LOGITS
     protect
    0.25
     treat
    0.23
    å°Ĭ
    0.23
     protecting
    0.22
     respect
    0.22
     fairly
    0.21
     protects
    0.21
    ä¿ĿæĬ¤
    0.20
    protect
    0.20
     Protect
    0.20
    Activations Density 0.102%

    No Known Activations