© 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. 18267
    Prev
    Next
    INDEX
    Explanations

    respectively, magnitude of vector

    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
    å¼ĢåıijåĪ©çĶ¨
    -0.11
    欢è¿İæĤ¨
    -0.10
    æľ¬ç½ij
    -0.10
    篱
    -0.09
    ì´ī
    -0.08
    ç¾ŀ
    -0.08
     Gregg
    -0.08
    CLUD
    -0.08
     inc
    -0.08
    Own
    -0.08
    POSITIVE LOGITS
    æijĦ
    0.11
    ä¸į容
    0.09
    念
    0.08
    orf
    0.08
    翼
    0.08
    é¢Ģ
    0.08
    caster
    0.08
    aro
    0.08
    æĭľè®¿
    0.08
    é³Į
    0.08
    Activations Density 0.001%

    No Known Activations