© 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. 59867
    Prev
    Next
    INDEX
    Explanations

    question starts with what

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

    questions that begin with a specific interrogative at the start of the user’s message.

    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
    <|begin_of_text|>
    -0.10
    _ASSUME
    -0.10
    idon
    -0.10
    оваÑĢи
    -0.09
    mé
    -0.09
     quot
    -0.09
    angl
    -0.09
    .utility
    -0.09
    roje
    -0.08
    OLON
    -0.08
    POSITIVE LOGITS
     kind
    0.15
     kinds
    0.13
    kind
    0.13
     else
    0.12
    else
    0.11
     types
    0.11
    æł·çļĦ
    0.10
     type
    0.10
     happens
    0.09
    zee
    0.09
    Activations Density 0.072%

    No Known Activations