© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20392
    Prev
    Next
    INDEX
    Explanations

    despite this

    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
    ä¸įä»ħ
    -0.10
    åIJĦ个çݯèĬĤ
    -0.09
    ç®Ģ缴
    -0.09
    ä¸įä»ħæĺ¯
    -0.09
    æľĢåIJĪéĢĤ
    -0.08
    åIJĦ个æĸ¹éĿ¢
    -0.08
    ç®Ģ缴æĺ¯
    -0.08
    æľĢ容æĺĵ
    -0.08
    大æ¦Ĥæĺ¯
    -0.08
    resse
    -0.08
    POSITIVE LOGITS
     ë¶Ī구íķĺê³ł
    0.11
    dice
    0.10
    æĢµ
    0.10
    ãģĵãģ¨ãģ«
    0.09
    _RC
    0.09
    æµ·éĩı
    0.09
    anja
    0.09
    éĶĢåĶ®æĶ¶åħ¥
    0.09
    woods
    0.09
    INDER
    0.09
    Activations Density 0.065%

    No Known Activations