© 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. 64203
    Prev
    Next
    INDEX
    Explanations

    refere Ḡao, potatoes underneath

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

    non‑English (especially Romance/Slavic) function words and common morphological endings, i.e., short connectors like conjunctions/prepositions and frequent suffixes within multilingual text.

    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
    ÂĢÂĢ
    -0.10
    \Dependency
    -0.08
    ******č\n
    -0.08
    krv
    -0.08
    printStats
    -0.07
     YYS
    -0.07
    åĽ°
    -0.07
    /epl
    -0.07
    /Dk
    -0.07
    POSITIVE LOGITS
     suce
    0.08
    ìĹŃìĭľ
    0.08
     индивидÑĥ
    0.08
    °
    0.07
     potentially
    0.07
    âĢī
    0.07
    à¹īาว
    0.07
     Glob
    0.07
     aforementioned
    0.07
     somew
    0.07
    Activations Density 1.059%

    No Known Activations