© 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. 61404
    Prev
    Next
    INDEX
    Explanations

    not enough, best, normal, baseline

    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
    epad
    -0.10
     Weed
    -0.10
    éĻ¢
    -0.09
    eree
    -0.09
    éĴĵ
    -0.09
    æĿł
    -0.08
    æ¢ħèĬ±
    -0.08
     Gos
    -0.08
    ewing
    -0.08
    æľĢåIJİä¸Ģ次
    -0.08
    POSITIVE LOGITS
    æĥħåĨµä¸ĭ
    0.11
    åīĬ
    0.09
    æľ¬èº«
    0.08
    геÑĢ
    0.08
    ÑĤек
    0.08
    æľ¬æĿ¥
    0.08
    å·²ç»ıæĺ¯
    0.08
    ä¸Ģ对
    0.08
    llib
    0.08
     favorable
    0.08
    Activations Density 0.040%

    No Known Activations