© 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. 2723
    Prev
    Next
    INDEX
    Explanations

    used as

    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
    asti
    -0.11
    åĭĭ
    -0.09
    åĤ¬
    -0.09
    OUSE
    -0.08
     gal
    -0.08
    çĽ¯çĿĢ
    -0.08
     Creed
    -0.08
     lonely
    -0.08
    ä¼¼
    -0.08
     Proud
    -0.08
    POSITIVE LOGITS
    èĢĮæĪIJ
    0.12
    ä¸įå¾Ĺå·²
    0.10
    æĿ¥çļĦ
    0.10
    à¸Ļำมา
    0.10
    ç»ĦæĪIJçļĦ
    0.09
    èĢĮæĿ¥
    0.09
    ernal
    0.09
    éĻįä½İæĪIJæľ¬
    0.09
    -derived
    0.09
    赶æĿ¥
    0.09
    Activations Density 0.307%

    No Known Activations