© 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. 24300
    Prev
    Next
    INDEX
    Explanations

    regret and repentance

    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
    å®ŀåľ¨
    -0.10
    çĦ¦
    -0.09
    åĽ¾ä¹¦é¦Ĩ
    -0.09
    ä¹Łè®¸
    -0.09
    .decorators
    -0.09
    æľ¬æĿ¥
    -0.08
    .EXTRA
    -0.08
    çŁ¥éģĵ
    -0.08
     bod
    -0.08
    ICO
    -0.08
    POSITIVE LOGITS
    åIJİæĤĶ
    0.22
    æĤĶ
    0.17
    éģĹæĨ¾
    0.17
     regret
    0.14
     regrets
    0.13
    åĴİ
    0.12
    æĨ¾
    0.12
    æĩĬ
    0.11
    éĶĻè¿ĩ
    0.11
    æĢ¨
    0.11
    Activations Density 0.026%

    No Known Activations