© 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. 57578
    Prev
    Next
    INDEX
    Explanations

    statistics and structure

    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
    æįIJ
    -0.09
    èĭĽ
    -0.09
    æ¸ħåįķ
    -0.08
    ä¸Ģ线
    -0.08
    æĴ¥
    -0.08
    弯
    -0.08
    .listFiles
    -0.08
    åħ¨å¤©
    -0.08
    èĮĥ
    -0.08
    é£İæ°´
    -0.08
    POSITIVE LOGITS
    å¿«æĬ¥
    0.09
    åªĴä»ĭ
    0.09
    è¦ģçŁ¥éģĵ
    0.09
     workforce
    0.09
    æľ¬æ¬¡
    0.09
    馥
    0.08
    æŃ¤æ¬¡
    0.08
    丼
    0.08
    awe
    0.08
    åıĺåĬ¨
    0.08
    Activations Density 0.037%

    No Known Activations