© 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. 33033
    Prev
    Next
    INDEX
    Explanations

    "by" or "for"

    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
    èĩªçIJĨ
    -0.08
    èijµ
    -0.08
    дей
    -0.08
    \Object
    -0.08
    meth
    -0.08
    absolute
    -0.08
     Trustees
    -0.08
    èĽĻ
    -0.08
    ÑģÑĭл
    -0.08
    /../
    -0.08
    POSITIVE LOGITS
    opot
    0.11
    inc
    0.10
     mainstream
    0.10
    éĤ£åıª
    0.09
    Inc
    0.09
    gor
    0.09
    æīįèĥ½å¤Ł
    0.09
    主æµģ
    0.09
    主导
    0.09
    ä»»ä½ķ
    0.09
    Activations Density 0.009%

    No Known Activations