© 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. 49337
    Prev
    Next
    INDEX
    Explanations

    explanation why

    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
    解æķ£
    -0.09
    社
    -0.08
    ipe
    -0.08
    imat
    -0.08
    é»ĺ
    -0.08
    енн
    -0.08
    ä¸ĢçĽ´åľ¨
    -0.08
    æĬ¥éħ¬
    -0.08
     purity
    -0.07
    èĪ¬
    -0.07
    POSITIVE LOGITS
    åİŁåĽł
    0.23
    çIJĨçĶ±
    0.20
    çļĦåİŁåĽł
    0.19
     reasons
    0.19
     reason
    0.18
     ìĿ´ìľł
    0.17
    reason
    0.17
     Reasons
    0.17
    çļĦçIJĨçĶ±
    0.16
    åİŁåĽłæĺ¯
    0.16
    Activations Density 0.161%

    No Known Activations