© 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. 59004
    Prev
    Next
    INDEX
    Explanations

    duration and recency

    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
    åľ¨æł¡
    -0.10
    Lifetime
    -0.09
    ä¸ĢçĶŁ
    -0.09
     lifetime
    -0.09
    åįģ大
    -0.08
     formerly
    -0.08
    âķ°
    -0.08
     sez
    -0.08
    éĻ¤å¤ķ
    -0.08
    nit
    -0.08
    POSITIVE LOGITS
    å·²ç»ıæľī
    0.16
    以æĿ¥
    0.14
    ä¸Ģå¹´å¤ļ
    0.12
    ä»İåݻ年
    0.12
     already
    0.12
    就已ç»ı
    0.11
    å·²ç»ıåľ¨
    0.11
    å·²ç»ı
    0.11
    å·²ç»ıå¼Ģå§ĭ
    0.11
    å¦Ļ
    0.10
    Activations Density 0.114%

    No Known Activations