© 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. 27690
    Prev
    Next
    INDEX
    Explanations

    quoted phrases

    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
    åľ¨éĤ£éĩĮ
    -0.10
    меÑģ
    -0.08
    两çĤ¹
    -0.08
    è´º
    -0.08
    æŶåĢĻ
    -0.08
     باÙĦØ£
    -0.08
    пÑĢе
    -0.08
    ä¸īçĤ¹
    -0.08
    ustr
    -0.08
    .Static
    -0.08
    POSITIVE LOGITS
    å¥
    0.09
    awi
    0.09
    说å®Į
    0.09
    æīĭä¸Ń
    0.08
    HELL
    0.08
    è¾ĥå°ı
    0.08
    èIJ½åħ¥
    0.08
    ċ
    0.08
    羨
    0.08
     зло
    0.08
    Activations Density 0.215%

    No Known Activations