© 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. 23574
    Prev
    Next
    INDEX
    Explanations

    use non-fat or regular

    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
    åĿ¨
    -0.09
    åŃIJåŃĻ
    -0.09
    åłµ
    -0.09
    ä¸įå¿ĺ
    -0.08
    terr
    -0.08
    èѦåĬ¡
    -0.08
    noxious
    -0.08
    èĬ±å¼Ģ
    -0.08
    èī
    -0.08
    rels
    -0.08
    POSITIVE LOGITS
    æľĢ好æĺ¯
    0.13
    åĵģçīĮ
    0.10
    é«ĺè´¨éĩı
    0.10
    ä¸ĵéĹ¨ä¸º
    0.10
     ones
    0.09
     brand
    0.09
    åĵģç§į
    0.09
    åŀĭåı·
    0.09
    abr
    0.09
    è¾ĥ好çļĦ
    0.09
    Activations Density 0.098%

    No Known Activations