© 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. 17441
    Prev
    Next
    INDEX
    Explanations

    pro- prefixes

    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
    geber
    -0.10
    ÐŀÑģ
    -0.09
    鹫
    -0.09
     parish
    -0.08
    çĽĨåľ°
    -0.08
    ian
    -0.08
    åijĺ
    -0.08
    樱èĬ±
    -0.08
     setHidden
    -0.08
    èģĮåĬ¡
    -0.08
    POSITIVE LOGITS
    æŃ¤äºº
    0.11
    BN
    0.10
    éĴı
    0.09
    flow
    0.09
    SN
    0.09
    _SN
    0.08
    åĢ¾
    0.08
    éĻ¤éĿŀ
    0.08
    ÂŃt
    0.08
    pped
    0.08
    Activations Density 0.030%

    No Known Activations