© 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. 53693
    Prev
    Next
    INDEX
    Explanations

    into law or action or effect

    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
    ableView
    -0.09
    æĺ¯éĿŀ
    -0.09
    /small
    -0.08
    rush
    -0.08
     Ingen
    -0.08
    _undo
    -0.08
    åĵ¡
    -0.08
    æĶ¶è´¹æłĩåĩĨ
    -0.08
    _REFRESH
    -0.08
    sticky
    -0.08
    POSITIVE LOGITS
    姻
    0.09
    æĮĤéĴ©
    0.08
    便åı¯
    0.08
    æ²Ł
    0.08
    ocket
    0.08
    åĩºè·¯
    0.08
     Simmons
    0.08
    itar
    0.08
    æķ´é¡¿
    0.08
    èĤ¤
    0.08
    Activations Density 0.005%

    No Known Activations