© 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. 6447
    Prev
    Next
    INDEX
    Explanations

    ending or breaking 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
    æĿ¥ä¸įåıĬ
    -0.09
    urd
    -0.09
    urally
    -0.09
    åĽ½ç½ij
    -0.09
    éļĹ
    -0.09
    ahead
    -0.09
    对å¾ħ
    -0.09
    è°¯
    -0.08
    cul
    -0.08
    ERN
    -0.08
    POSITIVE LOGITS
    æīĵçł´
    0.18
    æīĵçł´äºĨ
    0.15
    Breaking
    0.10
    çªģçł´
    0.10
    ç»ĵæĿŁäºĨ
    0.10
    çł´
    0.09
    ç¾ģ
    0.09
    åIJĪä½ľåħ³ç³»
    0.09
     ÑĩиÑģл
    0.09
    æ¡İ
    0.08
    Activations Density 0.023%

    No Known Activations