© 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. 21116
    Prev
    Next
    INDEX
    Explanations

    offensive, harmful, derogatory

    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
     unjust
    -0.12
    å£
    -0.11
     injust
    -0.10
     skeptic
    -0.10
     injustice
    -0.09
     Ung
    -0.09
     Kem
    -0.09
     Credential
    -0.09
    ç«ľ
    -0.09
    udo
    -0.08
    POSITIVE LOGITS
     offensive
    0.37
     offense
    0.32
     Offensive
    0.31
     sensitivity
    0.28
     offence
    0.26
     hurt
    0.26
    Insensitive
    0.23
     offend
    0.23
     offended
    0.22
    ensitivity
    0.21
    Activations Density 0.208%

    No Known Activations