© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40263
    Prev
    Next
    INDEX
    Explanations

    protection and safety

    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
    Negative Logits
    asser
    -0.10
    åħ´èĩ´
    -0.09
    ä¸Ģç¬Ķ
    -0.09
    rette
    -0.08
    _requires
    -0.08
    rosso
    -0.08
    led
    -0.08
    antro
    -0.08
    oola
    -0.08
    æĹłéĻIJ
    -0.08
    POSITIVE LOGITS
    ä¿ĿæĬ¤
    0.15
    ä¿ĿèŃ·
    0.13
     against
    0.13
     заÑīиÑĤ
    0.12
    éĺ²æĬ¤
    0.12
     Protect
    0.12
     protects
    0.12
    ä¿¡æģ¯å®īåħ¨
    0.12
    ä½ıäºĨ
    0.11
     shield
    0.11
    Activations Density 0.109%

    No Known Activations