© 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. 27765
    Prev
    Next
    INDEX
    Explanations

    matter, important, or difference

    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
    å¨ģ
    -0.10
    éģĤ
    -0.09
    æį§
    -0.09
    åį«
    -0.09
    allen
    -0.09
    æģ©
    -0.09
    åĪ°ä½į
    -0.09
    ä¸ĢåĪ»
    -0.09
    æīĵåĩº
    -0.09
    æĭIJ
    -0.09
    POSITIVE LOGITS
    æĹłæīĢè°ĵ
    0.17
    ä¸įåľ¨ä¹İ
    0.14
    æĹłåħ³
    0.13
    åıªè¦ģæĺ¯
    0.12
     mattered
    0.12
     irrelevant
    0.11
    ä¸į管
    0.11
    次è¦ģ
    0.11
     Matter
    0.11
    åıªè¦ģ
    0.11
    Activations Density 0.068%

    No Known Activations