© 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. 1518
    Prev
    Next
    INDEX
    Explanations

    plan attacks, use strategies

    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
    è¿Ļä¹Ī
    -0.09
    è®®
    -0.08
    å¦ĤæѤ
    -0.08
     yard
    -0.08
    åı¦å¤ĸä¸Ģ个
    -0.08
    ç¥ŀå¥ĩ
    -0.07
    èļĮ
    -0.07
     mim
    -0.07
    åı¦ä¸Ģç§į
    -0.07
     Bound
    -0.07
    POSITIVE LOGITS
    æĪĺçķ¥
    0.13
    çŃĸçķ¥
    0.13
    æĹ¢è¦ģ
    0.12
     strategy
    0.11
    .strategy
    0.10
     Strategy
    0.10
     strateg
    0.10
    æĪĺæľ¯
    0.10
     strategic
    0.09
    çļĦæĪĺçķ¥
    0.09
    Activations Density 0.047%

    No Known Activations