© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50587
    Prev
    Next
    INDEX
    Explanations

    -on place 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
    åħĭåĪ¶
    -0.09
    idot
    -0.09
    hu
    -0.09
    æijĨåľ¨
    -0.08
    å¼ĺ
    -0.08
    reich
    -0.08
     Trim
    -0.08
    åĹħ
    -0.08
    åĿIJä¸ĭ
    -0.08
     ÑĥÑģили
    -0.08
    POSITIVE LOGITS
    ç¦ıçī¹
    0.09
    驹
    0.09
     näch
    0.09
     Thái
    0.09
    ï¸
    0.09
    æĭĽèĤ¡
    0.08
    rub
    0.08
    +-+-
    0.08
    æłİ
    0.08
    .NoSuch
    0.08
    Activations Density 0.113%

    No Known Activations