© 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. 23941
    Prev
    Next
    INDEX
    Explanations

    technical prefixes and suffixes

    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
    utzer
    -0.10
    /Branch
    -0.09
    飧
    -0.09
    åIJİ代
    -0.09
    /feed
    -0.09
    ruc
    -0.09
    dorf
    -0.08
    å¨ģåIJį
    -0.08
    orem
    -0.08
     ÙħÙĪ
    -0.08
    POSITIVE LOGITS
    èµĢ
    0.08
    /or
    0.08
    æ¯ķ
    0.08
    ypy
    0.08
    对æĪijæĿ¥è¯´
    0.07
    _ic
    0.07
    æ¢Ń
    0.07
    æĭĵ
    0.07
    ç¾°
    0.07
    å¸ĥæľĹ
    0.07
    Activations Density 1.702%

    No Known Activations