© 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. 32635
    Prev
    Next
    INDEX
    Explanations

    checks and balances

    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
    ä¹ĭæīĢ
    -0.10
     annex
    -0.08
    ä¸Ģåı£
    -0.08
     обÑģлÑĥж
    -0.08
    åij¨è½¬
    -0.08
    é¤IJ
    -0.08
    æĺ¯éĿŀ
    -0.08
    anger
    -0.08
    Pros
    -0.08
    [opt
    -0.08
    POSITIVE LOGITS
    çĽijçĿ£
    0.20
    çĽijçĿ£æ£ĢæŁ¥
    0.16
    åĪ¶çº¦
    0.15
    æľºåĪ¶
    0.15
     oversight
    0.14
    çĽ£
    0.13
     Oversight
    0.13
     accountability
    0.13
    çĽijçĿ£ç®¡çIJĨ
    0.12
     supervision
    0.12
    Activations Density 0.054%

    No Known Activations