© 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. 63512
    Prev
    Next
    INDEX
    Explanations

    race to finish, who is stronger, compete in

    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
     spol
    -0.10
     scr
    -0.09
    swire
    -0.09
    846
    -0.09
    zon
    -0.09
    mere
    -0.08
     )\n\n\n\n\n\n\n\n
    -0.08
    lied
    -0.08
     Ja
    -0.08
    _finalize
    -0.08
    POSITIVE LOGITS
    è°ģ
    0.17
     who
    0.15
    who
    0.14
    誰
    0.13
     competing
    0.13
    à¹ĥà¸Ħร
    0.12
     whose
    0.12
     trying
    0.11
     kdo
    0.11
    競
    0.11
    Activations Density 0.072%

    No Known Activations