© 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. 11372
    Prev
    Next
    INDEX
    Explanations

    stomp, hop, tread

    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
    ç¼£
    -0.09
    wares
    -0.09
    umb
    -0.09
    家ä½ı
    -0.09
     ÑĢаÑģÑģ
    -0.09
    ãĤ§
    -0.09
    åı¯éĿłçļĦ
    -0.09
    áºŃy
    -0.08
    opr
    -0.08
    weet
    -0.08
    POSITIVE LOGITS
    æ¶İ
    0.11
    è·µ
    0.10
    åĬ¨çĶŁæĪIJ
    0.09
    pon
    0.09
    æĹłæķ°æ¬¡
    0.09
    è¸ı
    0.09
    è¿
    0.09
    chod
    0.09
    å¤ļåĬŁèĥ½
    0.09
    è¸ī
    0.09
    Activations Density 0.006%

    No Known Activations