© 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. 21800
    Prev
    Next
    INDEX
    Explanations

    winning or losing streaks

    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
    èĦĨ
    -0.10
    croft
    -0.09
    åĨ·éĿĻ
    -0.09
    举起
    -0.09
    èĤĸ
    -0.09
    inc
    -0.09
    好ä¸į好
    -0.08
    åľ°åĮº
    -0.08
    åijĹ
    -0.08
    openh
    -0.08
    POSITIVE LOGITS
    çļĦ妻åŃIJ
    0.09
    ess
    0.09
     Bauer
    0.09
    æ³·
    0.09
    æ¥ļ
    0.09
    бÑĭ
    0.08
    ä¸ĢéĿ¢
    0.08
    æĶĢ
    0.08
    æĶĢçĻ»
    0.08
    zas
    0.08
    Activations Density 0.010%

    No Known Activations