© 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. 29659
    Prev
    Next
    INDEX
    Explanations

    have been / have won

    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
    æĪij们éĥ½
    -0.12
    åĴ±
    -0.11
    该éĻ¢
    -0.10
    åѦçĶŁä»¬
    -0.10
    åĴĮå®ĮåĸĦ
    -0.10
    å§Ĭ妹
    -0.10
    æįħ
    -0.09
    ä¹Łæĺ¯ä¸Ģ个
    -0.09
    larımız
    -0.09
    åĽ½å®¶éĺŁ
    -0.09
    POSITIVE LOGITS
     themselves
    0.19
     Ñģами
    0.10
     имеÑİÑĤ
    0.10
    ä½Ĩä»ĸ们
    0.09
    éĽĨä½ĵ
    0.09
    åĽłä¸ºä»ĸ们
    0.09
    amp
    0.09
    æľīä¸Ģä½į
    0.09
    ä¸Ģä½į
    0.09
    ä¸Ģ个个
    0.08
    Activations Density 0.077%

    No Known Activations