© 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. 19224
    Prev
    Next
    INDEX
    Explanations

    in the form of

    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
    ç¿ģ
    -0.09
    å°Ĭ
    -0.09
    èµ·æĿ¥
    -0.09
    åĤ¬
    -0.09
    åŁł
    -0.08
    èħ´
    -0.08
     grö
    -0.08
    andr
    -0.08
    _authenticated
    -0.08
    稿
    -0.08
    POSITIVE LOGITS
    å½¢å¼ı
    0.21
    çļĦå½¢å¼ı
    0.19
     forme
    0.14
     form
    0.14
     dạng
    0.13
     виде
    0.13
     ÑĦоÑĢма
    0.13
     forma
    0.13
     ÑĦоÑĢме
    0.13
     íĺķíĥľ
    0.13
    Activations Density 0.112%

    No Known Activations