© 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. 16942
    Prev
    Next
    INDEX
    Explanations

    priority

    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
    jour
    -0.09
    æĸ°ä¸ļæĢģ
    -0.09
    íĺ¹
    -0.09
    æģ¸
    -0.09
     ÑĤоже
    -0.08
    çİĦ
    -0.08
    ä¸įä»ħè¦ģ
    -0.08
    èł¡
    -0.08
    å¨ģåĬĽ
    -0.08
     rfl
    -0.08
    POSITIVE LOGITS
    åħ¶æ¬¡
    0.11
    ä¼ĺåħĪ
    0.11
    第ä¸Ģä½į
    0.10
    åħ¶æ¬¡æĺ¯
    0.10
    é¦ĸè¦ģ
    0.10
    é¦ĸä½į
    0.10
     priority
    0.09
     Prim
    0.09
     prim
    0.09
    -secondary
    0.09
    Activations Density 0.048%

    No Known Activations