© 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. 29803
    Prev
    Next
    INDEX
    Explanations

    over the course 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
    æĻ°
    -0.10
    å±Ĥåĩº
    -0.09
    èĩªæĿ¥
    -0.09
    ìŀł
    -0.09
    ç¬ı
    -0.09
    æŶè£ħ
    -0.09
    ê¹IJ
    -0.09
    èĪĬ
    -0.09
    è¾Ĥ
    -0.09
    æľīä¸Ģç§į
    -0.08
    POSITIVE LOGITS
    æľĽåİ»
    0.11
    ä¸Ģ个å¤ļ
    0.10
    åįĬ个
    0.09
    ned
    0.09
    çķ¸
    0.08
    ä¸ļ
    0.08
    nings
    0.08
     обÑĥÑĩениÑı
    0.08
    å¾Ģ
    0.08
    å¼ıçļĦ
    0.08
    Activations Density 0.018%

    No Known Activations