© 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. 909
    Prev
    Next
    INDEX
    Explanations

    too often / too much / too many

    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
    ãģĬãģĭ
    -0.12
    ç¬ijæĦı
    -0.10
    缴éĢļ车
    -0.10
    æĪijæīį
    -0.09
    级以ä¸Ĭ
    -0.09
    åŃĹåı·
    -0.09
    毫æĹłçĸijéĹ®
    -0.09
    è¿ĺæĥ³
    -0.09
    åľ¨å®¶
    -0.08
    æĿ¡çº¦
    -0.08
    POSITIVE LOGITS
    å¾Ģå¾Ģ
    0.18
     often
    0.16
    å¾Ģå¾Ģæĺ¯
    0.14
     souvent
    0.14
    Often
    0.14
     Often
    0.14
    often
    0.13
     ÑĩаÑģÑĤо
    0.12
    æľīæŶ
    0.12
    常常
    0.12
    Activations Density 0.050%

    No Known Activations