© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21844
    Prev
    Next
    INDEX
    Explanations

    in all

    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
    æĸ¹å¼ıè¿Ľè¡Į
    -0.09
    éķ¿çĽ¸
    -0.09
    .sax
    -0.09
    é«ĺä½į
    -0.09
    ä»ģ
    -0.08
    arding
    -0.08
    身边çļĦ
    -0.08
    ä¹Łä¸įæķ¢
    -0.08
    upply
    -0.08
    æ··æ·Ĩ
    -0.08
    POSITIVE LOGITS
    æĢ»åħ±
    0.14
    åħ±æľī
    0.13
    ä¸Ģåħ±
    0.12
    åħ±è®¡
    0.10
     altogether
    0.10
    æĢ»è®¡
    0.10
    åħ±
    0.10
     вÑģего
    0.10
    æĢ»ä½ĵ
    0.10
     entirety
    0.10
    Activations Density 0.078%

    No Known Activations