© 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. 20900
    Prev
    Next
    INDEX
    Explanations

    until after

    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
    并äºİ
    -0.10
    æıIJæĹ©
    -0.09
    atty
    -0.09
    忽
    -0.09
    ongan
    -0.09
    etical
    -0.09
    ç»ŃèĪª
    -0.09
    è¥Ħ
    -0.09
    ä¹ŁæĽ¾
    -0.09
    ä¹ĭç¾İ
    -0.09
    POSITIVE LOGITS
     nor
    0.10
    ä¸ī个æĸ¹éĿ¢
    0.10
    鸽
    0.09
    /from
    0.09
    ä¸ĩåĪĨ
    0.08
    å¤ļå±Ĥ次
    0.08
    åīįåĪĹ
    0.08
    /as
    0.08
    å¾Ĺä½ı
    0.08
    åIJİéĿ¢
    0.08
    Activations Density 0.025%

    No Known Activations