© 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. 23079
    Prev
    Next
    INDEX
    Explanations

    deterioration

    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
    ammen
    -0.10
    queda
    -0.09
    stay
    -0.09
    ä¸įä¸Ĭ
    -0.09
    èļ¤
    -0.09
    atically
    -0.09
    è¹Ī
    -0.08
    ãģĤãģ£ãģŁ
    -0.08
    ı
    -0.08
    åĸĿæ°´
    -0.08
    POSITIVE LOGITS
     tá»ij
    0.11
    æľįåĬ¡è´¨éĩı
    0.10
    çĶŁæ´»è´¨éĩı
    0.10
    kop
    0.10
    /dist
    0.09
    ember
    0.09
    ä¸ĢåĽ¢
    0.09
    ìĭľíĤ¤
    0.09
    çĬ¶åĨµ
    0.09
    wyn
    0.09
    Activations Density 0.043%

    No Known Activations