© 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. 18528
    Prev
    Next
    INDEX
    Explanations

    from scratch

    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
    é©·
    -0.10
    泡泡
    -0.10
    ç½²
    -0.10
    ODO
    -0.09
    溢
    -0.09
    ULE
    -0.09
    宿
    -0.09
    èĹ»
    -0.09
    YLE
    -0.09
    æŃª
    -0.09
    POSITIVE LOGITS
    scratch
    0.11
     scratch
    0.11
    åºķå±Ĥ
    0.10
    åħ¨æĸ°
    0.10
     ground
    0.09
    æݧèĤ¡éĽĨåĽ¢
    0.09
    åľ°éĿ¢
    0.09
    åŁºæľ¬éĿ¢
    0.09
    åħ¨æĸ°çļĦ
    0.09
    éĩįæĸ°
    0.09
    Activations Density 0.048%

    No Known Activations