© 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. 21948
    Prev
    Next
    INDEX
    Explanations

    taking the reins of

    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
    åģ¥åħ¨
    -0.11
    工夫
    -0.10
    åĬŁå¤«
    -0.09
     باÙĦÙħ
    -0.09
    ulan
    -0.09
    羣çĪ±
    -0.09
    ä¸ŃåĽ½è¶³çIJĥ
    -0.08
     Mär
    -0.08
     stre
    -0.08
    éģ½
    -0.08
    POSITIVE LOGITS
    ấn
    0.10
    pons
    0.10
    èĻ¬
    0.09
    æİĮ
    0.09
    è¾Ķ
    0.09
    çļĦ责任
    0.09
    大èĥĨ
    0.09
     Riley
    0.09
    çĶ¨æīĭ
    0.09
    inis
    0.08
    Activations Density 0.020%

    No Known Activations