© 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. 49789
    Prev
    Next
    INDEX
    Explanations

    speaker references

    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´ĸ
    -0.09
    åĬ¬
    -0.09
    afen
    -0.09
    åıijæĮ¥äºĨ
    -0.09
    è½°
    -0.08
    dap
    -0.08
    åįĬå¹´
    -0.08
    _hal
    -0.08
    好ä¸į好
    -0.08
     jud
    -0.08
    POSITIVE LOGITS
     magician
    0.13
    ãĥįãĤ¿
    0.10
    æł½
    0.10
    éŃĶæľ¯
    0.10
    表æ¼Ķ
    0.09
    åĩĭ
    0.09
    ä¸ĢåIJį
    0.08
    ç»ıåħ¸çļĦ
    0.08
    anela
    0.08
    äººåľ¨
    0.08
    Activations Density 0.037%

    No Known Activations