© 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. 57310
    Prev
    Next
    INDEX
    Explanations

    discount percentage

    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
    eneg
    -0.10
    管çIJĨéĥ¨éŨ
    -0.09
    æĪĺèĥľ
    -0.09
    çĮ¿
    -0.09
    鬻
    -0.09
    ãģ¾ãģķ
    -0.09
    agus
    -0.09
    :num
    -0.08
    marvin
    -0.08
    .handleSubmit
    -0.08
    POSITIVE LOGITS
    修身
    0.11
    æīĢæľī
    0.09
    æīĢæľīçļĦ
    0.09
    åįĬæľĪ
    0.09
     Bauer
    0.08
    好èݱåĿŀ
    0.08
    4
    0.08
    æī«
    0.08
    ä¸ĢåĪĩ
    0.08
    ä¸ĢéĿ¢
    0.08
    Activations Density 0.005%

    No Known Activations