© 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. 22665
    Prev
    Next
    INDEX
    Explanations

    by, via, using, through, on, with

    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
    erot
    -0.08
    ä¸Ģäºĭ
    -0.08
    ç»Ħç»ĩå®ŀæĸ½
    -0.08
    ë°¸
    -0.08
    ünde
    -0.08
    éĤ£ä½į
    -0.08
    仨
    -0.08
    鲧
    -0.08
    entina
    -0.08
    ewriter
    -0.08
    POSITIVE LOGITS
    积
    0.09
    èĬĿ
    0.08
    æ¨Ĺ
    0.08
    ç®Ģ便
    0.08
    uss
    0.08
    æĺı
    0.08
     Cornel
    0.08
    åºķå±Ĥ
    0.08
     желез
    0.08
    MESS
    0.08
    Activations Density 0.315%

    No Known Activations