© 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. 12256
    Prev
    Next
    INDEX
    Explanations

    penalty and punishment

    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
     Ñĩа
    -0.10
    èĬ±éĴ±
    -0.10
    adi
    -0.10
    éıĬ
    -0.09
    ä¾¿å®ľ
    -0.08
    /part
    -0.08
    obuf
    -0.08
    -ves
    -0.08
    inand
    -0.08
    /ion
    -0.08
    POSITIVE LOGITS
     sanctions
    0.10
    å¥ĸåĬ±
    0.10
    æݪæĸ½
    0.10
    夯
    0.09
    åĢ¬
    0.09
    å¤Ħç½ļ
    0.09
    æĥ©ç½ļ
    0.09
    è·Łä¸įä¸Ĭ
    0.09
    åIJİæŀľ
    0.09
     penalties
    0.09
    Activations Density 0.028%

    No Known Activations