© 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. 4108
    Prev
    Next
    INDEX
    Explanations

    over handle, avoid feeding, overcooking, overtrained

    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
    endale
    -0.09
     поÑĤÑĢеб
    -0.09
    èĦĨ
    -0.09
    æ¸ĹéĢı
    -0.08
     perme
    -0.08
    å¿«ä¹IJ
    -0.08
    gy
    -0.08
    .hw
    -0.08
    ä¸ŃåĽ½ä¼ģä¸ļ
    -0.08
    peri
    -0.08
    POSITIVE LOGITS
    ä¸Ģè§Ĵ
    0.10
    åľºåIJĪ
    0.09
    第ä¸Ģ天
    0.09
    uzzle
    0.09
     ****************************************************************************
    0.09
    æİĸ
    0.09
    sti
    0.08
    å¡ŀ
    0.08
    Ĥ¬
    0.08
    soup
    0.08
    Activations Density 0.036%

    No Known Activations