© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Llama3.3-70B-IT (Instruct)
    5. 50-RESID-POST-GF
    7. 34627
    Prev
    Next
    INDEX
    Explanations

    harmful answers

    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
     culprit
    -0.10
     Cunning
    -0.09
    lesia
    -0.09
     Disposable
    -0.09
     Coul
    -0.09
     habitual
    -0.09
     Ùģض
    -0.09
    (íģ¬ê¸°
    -0.08
    oud
    -0.08
    preneur
    -0.08
    POSITIVE LOGITS
     content
    0.16
     harm
    0.14
     output
    0.14
     Harm
    0.13
     Content
    0.12
     outputs
    0.12
     hurt
    0.12
     AI
    0.12
     safety
    0.12
     Chat
    0.12
    Activations Density 0.143%

    No Known Activations