© 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. 36650
    Prev
    Next
    INDEX
    Explanations

    toxicity is not well understood

    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
     tang
    -0.09
     toxin
    -0.09
     trap
    -0.08
     cott
    -0.08
     Canter
    -0.08
     pros
    -0.08
     fkk
    -0.08
     ><?
    -0.08
    etag
    -0.08
    (æĹ¥
    -0.08
    POSITIVE LOGITS
     LD
    0.21
     acute
    0.16
     toxicity
    0.15
    LD
    0.14
     routes
    0.13
     mut
    0.13
     toxic
    0.12
     Toxic
    0.12
    acute
    0.11
     ld
    0.11
    Activations Density 0.013%

    No Known Activations