© 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. 19102
    Prev
    Next
    INDEX
    Explanations

    having or had

    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
    éĹ®ä»ĸ
    -0.11
    è¿Ļ段æŶéĹ´
    -0.10
    帮ä»ĸ
    -0.10
    togroup
    -0.10
    è¿Ļ对
    -0.10
    agy
    -0.09
    acy
    -0.09
    é²ľèĬ±
    -0.09
    å¾ĭ
    -0.09
    æīĭèħķ
    -0.09
    POSITIVE LOGITS
     doctrine
    0.09
    å°Ĩèĩªå·±çļĦ
    0.09
    legate
    0.09
    AMP
    0.08
    ä¸įäºĨ
    0.08
    æĪĸå°Ĩ
    0.08
    å°Ĩåľ¨
    0.08
    _embed
    0.08
    ç¼ĺ
    0.08
    curring
    0.08
    Activations Density 0.308%

    No Known Activations