© 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. 31573
    Prev
    Next
    INDEX
    Explanations

    first and second degree

    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
    åīįåIJİ
    -0.09
    æŁIJäºĽ
    -0.09
    ä¸įå¾Ĺä¸į
    -0.09
    GLOBALS
    -0.09
     sö
    -0.08
    äºĨä¸ĢäºĽ
    -0.08
    .pix
    -0.08
    washer
    -0.08
    panse
    -0.08
    ä¸įå¾Ĺå·²
    -0.08
    POSITIVE LOGITS
     type
    0.12
    æĴ¸
    0.10
    ç±»åŀĭ
    0.10
     mum
    0.10
    Assignable
    0.09
    cht
    0.09
    TYPE
    0.09
    SSF
    0.09
    _gradients
    0.09
    æ·Ģ
    0.08
    Activations Density 0.050%

    No Known Activations