© 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. 12813
    Prev
    Next
    INDEX
    Explanations

    command or instruction sequence

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

    document-structure and formatting cues, especially section headers, list markers, and sentence/phrase boundary punctuation.

    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
     )\n\n\n\n\n\n\n\n
    -0.09
     [âĢ¦
    -0.08
    ager
    -0.07
     pupper
    -0.07
     NotImplemented
    -0.07
     ...\n\n\n\n
    -0.07
    ly
    -0.07
     :\n\n\n\n
    -0.07
     Adams
    -0.07
     Gaz
    -0.07
    POSITIVE LOGITS
    ¶Į
    0.11
    ¦æĥħ
    0.11
    ĥ½
    0.11
     -*-č\n
    0.10
     dÃŃl
    0.08
     arada
    0.08
    .Formatter
    0.08
    DonaldTrump
    0.08
    _tokenize
    0.08
    exo
    0.08
    Activations Density 1.370%

    No Known Activations