© 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. 36888
    Prev
    Next
    INDEX
    Explanations

    used in or for

    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
    åºĶçĶ¨
    -0.10
    aside
    -0.09
    æīĭä¸ŃçļĦ
    -0.09
    å¾Ĺä½ı
    -0.09
    çĶ¨æīĭ
    -0.09
    åºĶçĶ¨æŁ¥çľĭ
    -0.09
    好è¿IJ
    -0.08
    æīĭåĬ¨
    -0.08
    绪
    -0.08
    被èªī
    -0.08
    POSITIVE LOGITS
    à¹ĥà¸Ļà¸ģาร
    0.13
    efault
    0.11
    RITE
    0.10
    çī¹åĪ«å£°æĺİ
    0.10
    æĮĿ
    0.09
    colm
    0.09
    ollen
    0.09
    MBER
    0.09
     noqa
    0.09
    hen
    0.08
    Activations Density 0.107%

    No Known Activations