© 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. 53619
    Prev
    Next
    INDEX
    Explanations

    edit this file

    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
    æŀķ
    -0.11
    æ¯Ļ
    -0.10
    åħļé£İ
    -0.10
    raph
    -0.09
    é¢Ĥ
    -0.09
    oose
    -0.09
    _sdk
    -0.09
    çĭİ
    -0.08
    ä¸ŃåĽ½æ¢¦
    -0.08
    ä¸įæĺ¯å¾Ī
    -0.08
    POSITIVE LOGITS
    _own
    0.09
    èĮ«çĦ¶
    0.09
    寡
    0.09
     swój
    0.09
     own
    0.09
     Morrow
    0.08
    stroy
    0.08
     ho
    0.08
    uco
    0.08
     Own
    0.08
    Activations Density 0.009%

    No Known Activations