© 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. 28209
    Prev
    Next
    INDEX
    Explanations

    numbers followed by units or specific words

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

    expressions of quantity or specification (especially numerals and range/variety markers) and clause-level modality/negation.

    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
    <::
    -0.09
    yen
    -0.09
     ALERT
    -0.08
     Carroll
    -0.08
    onta
    -0.08
    argar
    -0.08
    vak
    -0.08
     Jays
    -0.08
    tej
    -0.08
     Orient
    -0.08
    POSITIVE LOGITS
     Hogan
    0.09
     Sach
    0.08
    èĩ´
    0.08
    æª
    0.08
    wd
    0.08
     embod
    0.08
    vanced
    0.08
     presum
    0.08
     Smile
    0.08
    æ´¥
    0.08
    Activations Density 0.354%

    No Known Activations