© 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. 61645
    Prev
    Next
    INDEX
    Explanations

    ingredients and dishes

    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
    æ¶Īæŀģ
    -0.10
    éŀ®
    -0.10
    æģ¶åĮĸ
    -0.10
    åħ´æĹº
    -0.09
    åĤį
    -0.09
    躯
    -0.09
    æľĭåıĭ们
    -0.09
    éĺĶ
    -0.09
    èĩªæĿ¥æ°´
    -0.09
    ringe
    -0.09
    POSITIVE LOGITS
    大èĥĨ
    0.09
    ç²¾éĢī
    0.08
    èµ
    0.08
     theat
    0.08
     Impossible
    0.08
    åĮĹæĸĹ
    0.08
    .gg
    0.08
    带åĽŀ
    0.08
    ä¸įåŃĺåľ¨
    0.08
     impossible
    0.07
    Activations Density 0.030%

    No Known Activations