© 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. 571
    Prev
    Next
    INDEX
    Explanations

    consists of

    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
    æŀī
    -0.10
    款
    -0.10
    ogi
    -0.10
     наÑĪ
    -0.09
    å½¹
    -0.09
    rack
    -0.09
    åħ¶ä¸Ńæľī
    -0.09
    涯
    -0.09
    _tb
    -0.09
    ray
    -0.08
    POSITIVE LOGITS
    两个
    0.12
    ä¸Ģç³»åĪĹ
    0.10
    åĩłä¸ª
    0.10
     неÑģколÑĮкиÑħ
    0.09
    ä¸ī个
    0.09
    ä¸įä½ı
    0.09
    以ä¸ĭåĩłä¸ª
    0.09
     two
    0.09
    åľ°ä¸Ĭ
    0.08
    åĵª
    0.08
    Activations Density 0.146%

    No Known Activations