© 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. 2364
    Prev
    Next
    INDEX
    Explanations

    defects, fault, defective workmanship

    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
    çĶŁèĤ²
    -0.09
    éĻį
    -0.09
    ä¸Ģæ³¢
    -0.09
    天é¹ħ
    -0.08
    å¢ŀ
    -0.08
    weed
    -0.08
    o
    -0.08
     sandbox
    -0.08
    ouri
    -0.08
    ä¸Ńä»ĭ
    -0.08
    POSITIVE LOGITS
    agues
    0.10
    åĨħåľ¨
    0.10
    æľīä¸Ģä½į
    0.10
    .Duration
    0.10
    åįĬ导
    0.09
    åıijçİ°éĹ®é¢ĺ
    0.09
    ç»Ļä»ĸ们
    0.09
    èĤĦ
    0.09
    èĸª
    0.09
    窳
    0.09
    Activations Density 0.010%

    No Known Activations