© 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. 55098
    Prev
    Next
    INDEX
    Explanations

    previous, next, higher, downstream

    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
    åĩºèĩª
    -0.10
    åºĶæĢ¥ç®¡çIJĨ
    -0.10
    æĸ°æĹ¶æľŁ
    -0.10
    åħ¶æīĢ
    -0.09
    oyer
    -0.09
    好äºĨ
    -0.09
    å¿ĥåºķ
    -0.09
    ystack
    -0.09
    èĶ
    -0.09
    åī©ä¸ĭ
    -0.09
    POSITIVE LOGITS
     generations
    0.11
    æĬĬæĪij
    0.09
     scr
    0.09
    对æĪij
    0.09
    ä¸ĢéĿ¢
    0.08
    çŀ»
    0.08
    å·§
    0.08
    auc
    0.08
    cas
    0.08
    鸡
    0.08
    Activations Density 0.174%

    No Known Activations