© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22070
    Prev
    Next
    INDEX
    Explanations

    next, followed by

    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
    æľĢåIJİä¸Ģ次
    -0.11
    иж
    -0.10
    æľĢç»Ī
    -0.10
    esen
    -0.09
    æľĢåIJİä¸Ģ
    -0.09
     ultimately
    -0.09
    çŁŃçŁŃ
    -0.08
    è¿Ļä¸ªåľ°åĽ¾
    -0.08
     полÑĥÑĩи
    -0.08
     Fortune
    -0.08
    POSITIVE LOGITS
    åħ¶æ¬¡æĺ¯
    0.16
    åħ¶æ¬¡
    0.16
    ä¾Ŀ次
    0.11
    æİ¥çĿĢ
    0.11
    Next
    0.10
     Next
    0.10
    åħ³åĪĩ
    0.10
    èĶ·
    0.09
    next
    0.09
    -next
    0.09
    Activations Density 0.042%

    No Known Activations