© 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. 61619
    Prev
    Next
    INDEX
    Explanations

    properly stored

    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
    ä¹IJ
    -0.10
    表æĢģ
    -0.09
    _RESERVED
    -0.09
    _COMPAT
    -0.09
    æĶ¶æķĽ
    -0.09
    å»ºåĽ½
    -0.09
    ä¹Łä¸įä¾ĭå¤ĸ
    -0.09
    èĨ³
    -0.09
    ä¾Ľå¤§å®¶
    -0.08
    _Release
    -0.08
    POSITIVE LOGITS
    mand
    0.09
    Interior
    0.08
    MAND
    0.08
    achel
    0.08
    åŀĽ
    0.07
    ç»ŀ
    0.07
    人éĻħ
    0.07
    lessly
    0.07
    çĶŁäº§æĢ»åĢ¼
    0.07
    miner
    0.07
    Activations Density 0.988%

    No Known Activations