© 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. 21809
    Prev
    Next
    INDEX
    Explanations

    | Muslim | laws | hypocrisy | conservative | Taliban | Sharia | attack | intolerance | repressed | values

    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
    éĢīæĭ
    -0.10
    _runtime
    -0.09
    æĹłç§ģ
    -0.09
    éĵ¨
    -0.08
    ẩu
    -0.08
    å¹³åİŁ
    -0.08
    èĩªæĿ¥
    -0.08
    umpt
    -0.08
     kla
    -0.08
    åݨæĪ¿
    -0.08
    POSITIVE LOGITS
    ä¿Ŀå®Ī
    0.10
     conservative
    0.10
    æ´ģ
    0.10
    åħ·
    0.10
     Spare
    0.10
    æī¼
    0.09
    墨
    0.09
    lash
    0.09
     intolerance
    0.09
    æĮĩ导
    0.09
    Activations Density 0.132%

    No Known Activations