© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 11224
    Prev
    Next
    INDEX
    Explanations

    expressing regret or wishing

    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
    ä¸ĩ象
    -0.09
    èĩ´
    -0.09
    è°Ħ
    -0.08
    ä¸ĢçĽ´æ²¡æľī
    -0.08
     expect
    -0.08
    åıįæĬĹ
    -0.08
    èµ·ä¼ı
    -0.08
    èī¯
    -0.08
    eway
    -0.08
     expectation
    -0.08
    POSITIVE LOGITS
    åIJİæĤĶ
    0.33
    æĩĬ
    0.22
     regret
    0.22
    æĤĶ
    0.21
     regrets
    0.18
    éģĹæĨ¾
    0.17
    äºĭåIJİ
    0.14
    å¿ı
    0.11
    åĽŀæĥ³
    0.11
    æĦŁåı¹
    0.11
    Activations Density 0.045%

    No Known Activations