© 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. 27343
    Prev
    Next
    INDEX
    Explanations

    wayward, way above, way too

    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
    maal
    -0.12
    nier
    -0.10
     Webseite
    -0.10
    unct
    -0.10
     Airways
    -0.09
    ses
    -0.09
    ique
    -0.09
    бÑĥдÑĮ
    -0.09
    avig
    -0.09
    pyx
    -0.09
    POSITIVE LOGITS
    ward
    0.30
    far
    0.20
    finding
    0.19
    fair
    0.15
    WARD
    0.15
    yy
    0.13
     thức
    0.13
    bill
    0.13
    fare
    0.13
    finder
    0.13
    Activations Density 0.057%

    No Known Activations