© 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. 32008
    Prev
    Next
    INDEX
    Explanations

    read later

    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
    /or
    -0.10
    ppard
    -0.09
    çĶ¨æĿ¥
    -0.08
    ä¸įå¼Ģ
    -0.08
    åħ®
    -0.08
    ->__
    -0.08
    宫
    -0.07
     Thames
    -0.07
    å¾Ī强
    -0.07
    ÑĥÑī
    -0.07
    POSITIVE LOGITS
     sake
    0.14
    æĺ¯æľĢ好çļĦ
    0.11
    å®ŀåľ¨æĺ¯
    0.10
    åıĪä¸Ģ次
    0.10
    åĪ°æŶåĢĻ
    0.09
    è´»
    0.09
    bart
    0.09
    æĹ©æĹ¥
    0.09
     purposes
    0.09
    _unc
    0.09
    Activations Density 0.287%

    No Known Activations