© 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. 35422
    Prev
    Next
    INDEX
    Explanations

    plays double roles as

    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
    å§ijå¨ĺ
    -0.09
    å°ıå§IJ
    -0.09
    马æĿ¥
    -0.09
    igit
    -0.09
     piping
    -0.09
    æľ¯è¯Ń
    -0.08
    xda
    -0.08
     Morrow
    -0.08
    amina
    -0.08
    -animate
    -0.08
    POSITIVE LOGITS
    ä¿Ŀå®ī
    0.09
    æĸ°éĹ»è®°èĢħ
    0.09
    å¸®ä½ł
    0.09
    çĸ¤
    0.09
    å¹´æĬ¥
    0.08
    ä»ħä¾Ľ
    0.08
    寸
    0.08
    .removeChild
    0.08
    ninger
    0.08
    çIJĨæĻº
    0.08
    Activations Density 0.048%

    No Known Activations