© 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. 8243
    Prev
    Next
    INDEX
    Explanations

    names appearing in texts

    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
    upal
    -0.10
    ä¸įå½ĵ
    -0.10
    ä¾Ľ
    -0.09
    åįİ举
    -0.09
    aux
    -0.09
    _COMMIT
    -0.09
     اÙĦتش
    -0.08
    orm
    -0.08
    -hop
    -0.08
    åĨ°åĨ·
    -0.08
    POSITIVE LOGITS
    éĺ¿æĭī伯
    0.11
    æ²Ļæ¼ł
    0.10
    æ²Ļçī¹
    0.10
    ä¼Ĭæĸ¯åħ°
    0.10
    qed
    0.10
     SALE
    0.10
    å®ģæĦ¿
    0.10
     Sheikh
    0.10
    addin
    0.09
     Al
    0.09
    Activations Density 0.151%

    No Known Activations