© 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. 36966
    Prev
    Next
    INDEX
    Explanations

    sharpness or sharp angles

    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
    飫
    -0.10
    æĦĨ
    -0.09
    æ±Ĺæ°´
    -0.09
    马æĭī
    -0.09
    社ä¼ļèµĦæľ¬
    -0.08
    mpl
    -0.08
    zem
    -0.08
    å¹¼ç¨ļ
    -0.08
    uario
    -0.08
     heavy
    -0.08
    POSITIVE LOGITS
    ness
    0.15
    éĶIJ
    0.13
    sharp
    0.11
    éĶĭ
    0.11
    伶
    0.10
    æŀľæĸŃ
    0.10
     sharper
    0.10
    shoot
    0.10
    Sharp
    0.09
    cuts
    0.09
    Activations Density 0.023%

    No Known Activations