© 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. 38561
    Prev
    Next
    INDEX
    Explanations

    country or nation context

    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
    éĬ®
    -0.10
    çŃ
    -0.10
    atrice
    -0.09
    ä»İæĿ¥ä¸į
    -0.09
    æ¶Ī
    -0.09
    ä»İæĿ¥æ²¡
    -0.08
    _HOR
    -0.08
    壮
    -0.08
    ç»ıæµİä½ĵ
    -0.08
    æ°¸è¿ľä¸įä¼ļ
    -0.08
    POSITIVE LOGITS
    éĶij
    0.09
    expert
    0.09
    istic
    0.08
    ÑĢаÐ
    0.08
    åIJĦçķĮ
    0.08
    /exp
    0.08
    åIJĦ
    0.08
    èī¯
    0.08
    poon
    0.08
    beat
    0.08
    Activations Density 0.014%

    No Known Activations