© 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. 17187
    Prev
    Next
    INDEX
    Explanations

    makes someone a better

    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
     fare
    -0.10
    æ²IJæµ´
    -0.09
    ä¹ħäºĨ
    -0.09
    ä¹Łæĺ¯ä¸Ģç§į
    -0.09
    fare
    -0.09
    å¾ħ
    -0.09
    mouth
    -0.08
     обÑĭ
    -0.08
    ä¸Ģèīĺ
    -0.08
    (shared
    -0.08
    POSITIVE LOGITS
    éĿ¢çĽ®
    0.10
    羸
    0.10
    ä¸į对
    0.09
    空ä¸Ń
    0.09
    æĪª
    0.09
    åħ¨èĩªåĬ¨
    0.08
    åĵī
    0.08
     Worth
    0.08
    åij½
    0.08
    onomic
    0.08
    Activations Density 0.054%

    No Known Activations