© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8845
    Prev
    Next
    INDEX
    Explanations

    characters and series names

    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
    chai
    -0.10
    é½Ĭ
    -0.09
    çķ¦
    -0.09
    urous
    -0.09
    enth
    -0.09
    ög
    -0.08
    _THIS
    -0.08
    Ñģол
    -0.08
    è¯ĺ
    -0.08
    PKG
    -0.08
    POSITIVE LOGITS
    è¦ģ说
    0.09
    çļĦèĥĮåIJİ
    0.09
    éĨ´
    0.09
    ãģĵãģ¨
    0.09
    娱ä¹IJå¹³åı°
    0.09
    weit
    0.09
    çļĦåĪ°æĿ¥
    0.09
    EW
    0.08
    åī¥
    0.08
     EW
    0.08
    Activations Density 0.083%

    No Known Activations