© 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. 4626
    Prev
    Next
    INDEX
    Explanations

    most likely to

    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
    ayed
    -0.09
    è¿ĩ硬
    -0.09
    avid
    -0.08
    <S
    -0.08
    Ñıн
    -0.08
     Lap
    -0.08
    ä½ĵåĪ¶æľºåĪ¶
    -0.08
    NavigationView
    -0.08
    å¿Ļ
    -0.08
    è¶Ł
    -0.08
    POSITIVE LOGITS
    çķĪ
    0.11
    èģ¯çµ¡
    0.09
     mắn
    0.09
    æĮ¯åħ´
    0.09
    åĹ£
    0.09
    æĸĩæĺĮ
    0.08
    hood
    0.08
    æĭ¢
    0.08
    ÑĢаÑģÑĤ
    0.08
    à¸Ĺรà¸ĩ
    0.08
    Activations Density 0.054%

    No Known Activations