© 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. 18338
    Prev
    Next
    INDEX
    Explanations

    positive outcomes

    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
    å¿Ŀ
    -0.10
    æĪĺçķ¥æĢ§
    -0.10
    picker
    -0.10
    ç½®
    -0.09
     representative
    -0.09
    æĸ°åĵģ
    -0.09
    容å¿į
    -0.09
    çļĦæĦıä¹ī
    -0.09
    تØŃ
    -0.09
    æIJģ
    -0.09
    POSITIVE LOGITS
    æĪIJåĬŁ
    0.15
     proven
    0.14
    æĪIJåĬŁçļĦ
    0.14
    çļĦæĪIJåĬŁ
    0.14
     success
    0.12
    æĪIJæķĪ
    0.12
    ç»ıéªĮ
    0.12
     successful
    0.11
     proved
    0.11
    沿
    0.10
    Activations Density 0.112%

    No Known Activations