© 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. 27964
    Prev
    Next
    INDEX
    Explanations

    rejects or demands

    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
    ä¸įåĨįæĺ¯
    -0.10
    éĨĴ
    -0.09
    Stub
    -0.09
    轸
    -0.09
    ä¸įåıªæĺ¯
    -0.09
    ä¸įåı¯ç¼ºå°ij
    -0.09
    missing
    -0.08
    ç®´
    -0.08
    ä¸įæĸŃåľ°
    -0.08
    å¥
    -0.08
    POSITIVE LOGITS
     attempts
    0.14
    ä¼ģåĽ¾
    0.13
    è¯ķåĽ¾
    0.13
     requests
    0.12
     attempt
    0.12
    è¦ģæ±Ĥ
    0.11
     demands
    0.11
    æıIJè®®
    0.11
    æıIJåĩºçļĦ
    0.11
    _attempts
    0.10
    Activations Density 0.061%

    No Known Activations