© 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. 35202
    Prev
    Next
    INDEX
    Explanations

    hereby licensed or granted

    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
    带æĿ¥
    -0.08
    erase
    -0.08
     erase
    -0.08
    æľĭåıĭ们
    -0.08
    ç³Ĭ
    -0.08
    å¢ĻéĿ¢
    -0.08
    ал
    -0.08
    æĹłæĦı
    -0.07
    ewater
    -0.07
    DOWN
    -0.07
    POSITIVE LOGITS
    好çľĭ
    0.10
    Ùģت
    0.10
    çīĮ
    0.10
    å¨Ī
    0.10
    heit
    0.09
    .breakpoints
    0.09
    çIJ´
    0.09
    çŃ¹
    0.09
    éķ¯
    0.09
    adius
    0.09
    Activations Density 0.005%

    No Known Activations