© 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. 15805
    Prev
    Next
    INDEX
    Explanations

    do harm or anything

    np_acts-logits-general · gemini-2.5-flash-lite

    direct second-person address and infinitive “to”-verb constructions, i.e., statements aimed at “you” expressing capability or intent.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
     Rescue
    -0.09
    lev
    -0.09
    ä¼ij
    -0.08
     Levin
    -0.08
    lek
    -0.08
     impro
    -0.08
     sport
    -0.08
     DEFINE
    -0.08
     importantly
    -0.08
    úsqueda
    -0.08
    POSITIVE LOGITS
     Burn
    0.10
     burner
    0.10
     gle
    0.09
     hã
    0.09
     destruction
    0.09
    afort
    0.09
     aprove
    0.09
     lul
    0.09
     Sherman
    0.09
     fire
    0.09
    Activations Density 0.089%

    No Known Activations