© 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. 54115
    Prev
    Next
    INDEX
    Explanations

    survived

    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
     dõi
    -0.09
    gregate
    -0.09
    ottage
    -0.09
    pling
    -0.09
    DOWN
    -0.09
     Crosby
    -0.09
    resse
    -0.09
    sg
    -0.09
    åĭĩäºİ
    -0.09
    UCT
    -0.08
    POSITIVE LOGITS
    å¯ĴåĨ¬
    0.10
    çİĩ为
    0.09
    é¾Ļæ¹ĸ
    0.09
    ä¸ĢäºĮ
    0.09
    yun
    0.08
    é£İåIJ¹
    0.08
    ä¸Ģ段æŶéĹ´
    0.08
    ä¸Ģ段
    0.08
     vard
    0.08
    çݰ代社ä¼ļ
    0.08
    Activations Density 0.018%

    No Known Activations