© 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. 19455
    Prev
    Next
    INDEX
    Explanations

    shot wide

    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
    é²ľæĺİ
    -0.10
    è¡¥é½IJ
    -0.10
    ç»Ģ
    -0.10
    ä¸įåIJĥ
    -0.09
    nbr
    -0.09
    æłĩéħį
    -0.09
    çļĦæľĭåıĭ
    -0.09
    ä»Ļå¢ĥ
    -0.08
    è¿Ļ对
    -0.08
    好åIJ§
    -0.08
    POSITIVE LOGITS
    _past
    0.09
    åģı离
    0.09
    åĬ£
    0.08
     Past
    0.08
    æĿĸ
    0.08
    .Col
    0.08
    触åıĬ
    0.08
    _BIND
    0.08
    æľªèĥ½
    0.08
     past
    0.08
    Activations Density 0.022%

    No Known Activations