© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 18839
    Prev
    Next
    INDEX
    Explanations

    Romanian, Hungarian, prince, princess

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    igu
    -0.11
    âĿ®
    -0.10
    imbus
    -0.09
    寮
    -0.09
    æļ¹
    -0.09
    antino
    -0.09
    èįĶæŀĿ
    -0.09
    éģı
    -0.09
    åħĪè¿Ľ
    -0.08
    åIJ¹
    -0.08
    POSITIVE LOGITS
    æ±Łè¥¿çľģ
    0.09
     Hungarian
    0.09
    éļĶ
    0.09
    elm
    0.08
    choice
    0.08
     plein
    0.08
     MEP
    0.08
    ä¸įåĬ¨
    0.08
    è¥Ł
    0.08
    åįĹæĺĮ
    0.08
    Activations Density 0.283%

    No Known Activations