© 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. 23949
    Prev
    Next
    INDEX
    Explanations

    with or using

    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
    è·¨
    -0.09
    ette
    -0.09
    idd
    -0.08
    èIJ½åħ¥
    -0.08
    é¾Ħ
    -0.08
    utz
    -0.08
    沦为
    -0.08
    ä»»ä½ķ人éĥ½
    -0.08
    æĹ¶è¾°
    -0.08
    èIJ½åľ¨
    -0.08
    POSITIVE LOGITS
    çł¾
    0.10
    è®°å¾Ĺ
    0.09
    ä¿Ŀ驾
    0.09
    éĵ§
    0.09
    ãĤ¢ãĤ¯ãĤ»
    0.09
    马æĭī
    0.09
    vron
    0.09
    个å°ı
    0.09
     yourself
    0.09
    ADER
    0.09
    Activations Density 0.006%

    No Known Activations