© 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. 10704
    Prev
    Next
    INDEX
    Explanations

    about to VERB

    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
    æīĢåľ¨
    -0.11
    å½ĵæŶ
    -0.11
    åĢŃ
    -0.10
    æľ¬æĬ¥
    -0.10
    å¾Ģå¾Ģ
    -0.10
    è¿ijå¹´æĿ¥
    -0.10
    å¤ļå¹´æĿ¥
    -0.09
    åĵ¡
    -0.09
    o
    -0.09
    cheid
    -0.09
    POSITIVE LOGITS
    çĦ¶æĺ¯
    0.10
     Raz
    0.10
    /current
    0.09
    æĪĸå°Ĩ
    0.08
    ifi
    0.08
    ETY
    0.08
    arded
    0.08
    aneous
    0.08
    æĪIJ为ä¸Ģ个
    0.08
    erness
    0.08
    Activations Density 0.006%

    No Known Activations