© 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. 54279
    Prev
    Next
    INDEX
    Explanations

    something's own characteristics

    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
    两级
    -0.10
    æĶ¯
    -0.08
    黥
    -0.08
    æĪIJ为ä¸ŃåĽ½
    -0.08
    綦
    -0.08
    对çħ§
    -0.08
    ç©·
    -0.08
    mere
    -0.08
    åħ«åŃĹ
    -0.08
    secure
    -0.08
    POSITIVE LOGITS
    åIJĦèĩªçļĦ
    0.15
    çĭ¬çī¹çļĦ
    0.14
    åIJĦèĩª
    0.14
    çī¹çĤ¹
    0.14
    çĭ¬çī¹
    0.13
     unique
    0.13
     Ñĥника
    0.12
    æľīèĩªå·±çļĦ
    0.12
    Unique
    0.11
    unique
    0.11
    Activations Density 0.053%

    No Known Activations