© 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. 8436
    Prev
    Next
    INDEX
    Explanations

    > email quoting

    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
    èĿī
    -0.10
    çħ§æł·
    -0.09
     Balk
    -0.09
    转身
    -0.08
    çĹĺ
    -0.08
    æŁ¥çľĭåħ¨æĸĩ
    -0.08
    çļĦçIJĨçĶ±
    -0.08
    以èĩ³äºİ
    -0.08
     Tango
    -0.08
    .FR
    -0.08
    POSITIVE LOGITS
    .strict
    0.10
    说äºĨ
    0.09
    ä¸¥æł¼
    0.09
    Ñĥз
    0.09
    建议
    0.09
    åħ³äºİ
    0.09
    ä¸įæħİ
    0.09
    æıIJåĩºäºĨ
    0.08
    åĨĻçļĦ
    0.08
    (.)
    0.08
    Activations Density 0.055%

    No Known Activations