© 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. 62252
    Prev
    Next
    INDEX
    Explanations

    specifically il

    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
    æĪijè¿ĺ
    -0.13
    æĪij羣çļĦ
    -0.12
    æĪijè¿ĺæĺ¯
    -0.12
    ä¸ĭéĿ¢æĪij们
    -0.11
    ä¸ĭéĿ¢å°±
    -0.11
    æĪijä»¬ä¹Ł
    -0.11
    éķ¿æķĪæľºåĪ¶
    -0.11
    è¿ĺæľīä»Ģä¹Ī
    -0.10
    æĪij以为
    -0.10
    åľĺ
    -0.10
    POSITIVE LOGITS
    ä¸Ģæĸ¹éĿ¢
    0.10
    éĢĤæŶ
    0.08
     Chance
    0.08
    èĻ½çĦ¶
    0.07
    _opts
    0.07
     Obs
    0.07
    roach
    0.07
    ames
    0.07
    _wr
    0.07
    建ç«ĭ
    0.07
    Activations Density 0.071%

    No Known Activations