© 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. 47849
    Prev
    Next
    INDEX
    Explanations

    comparison using as

    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
    Ñģка
    -0.09
    éģº
    -0.09
    ears
    -0.09
    ä¸ĢæŃ¥ä¸ĢæŃ¥
    -0.09
    åģļåĩº
    -0.09
    è¿İæĿ¥
    -0.08
    èĥĮçĿĢ
    -0.08
    åīįæıIJæĺ¯
    -0.08
    æ°ijèѦ
    -0.08
    ä¸ĭåįĬ
    -0.08
    POSITIVE LOGITS
    -than
    0.11
    Hel
    0.10
    devil
    0.09
    ream
    0.09
    è¾Ī
    0.09
    èİ«è¿ĩäºİ
    0.09
    woods
    0.09
    hell
    0.09
     Hel
    0.08
    åłª
    0.08
    Activations Density 0.020%

    No Known Activations