© 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. 39982
    Prev
    Next
    INDEX
    Explanations

    subject or problem

    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
    å½¢æĪIJçļĦ
    -0.11
    ä¿ĥæĪIJ
    -0.09
    ynos
    -0.09
    ä¿ĥ使
    -0.09
    onis
    -0.09
    .cms
    -0.09
    èµĥ
    -0.09
    GAN
    -0.08
    建ç«ĭåģ¥åħ¨
    -0.08
    ç´§æİ¥çĿĢ
    -0.08
    POSITIVE LOGITS
    ç¿ĺ
    0.09
     Kra
    0.09
    åı²ä¸ĬæľĢ
    0.08
    å²³
    0.08
    æģķ
    0.08
    娱ä¹IJ
    0.08
    没å¿ħè¦ģ
    0.08
    ::<
    0.08
    uple
    0.08
     imper
    0.07
    Activations Density 0.223%

    No Known Activations