© 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. 11753
    Prev
    Next
    INDEX
    Explanations

    harmful or abusive behavior

    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
     sodom
    -0.16
     Lesbian
    -0.13
     homosexual
    -0.12
     masturbating
    -0.12
     Penis
    -0.12
     penis
    -0.12
     dildo
    -0.11
     homosex
    -0.11
     homosexuality
    -0.11
     lesbian
    -0.11
    POSITIVE LOGITS
     consent
    0.22
     Consent
    0.20
     consenting
    0.18
     cons
    0.16
     dungeon
    0.15
     fantasies
    0.14
     safety
    0.14
     fet
    0.14
     saf
    0.14
     humiliation
    0.12
    Activations Density 0.181%

    No Known Activations