© 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. 32212
    Prev
    Next
    INDEX
    Explanations

    distinguishes

    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
    è¿Ļæĸ¹éĿ¢
    -0.10
    åĩłåįĥ
    -0.10
    æľīä¿¡å¿ĥ
    -0.09
    rame
    -0.09
    è¯ļä¿¡
    -0.09
    æķ°åįĥ
    -0.08
    phis
    -0.08
    æľ¬ç½ijç«Ļ
    -0.08
    大å®Ĺ
    -0.08
    aza
    -0.08
    POSITIVE LOGITS
     dashes
    0.08
    éĨī
    0.08
    stre
    0.08
    ä¾Ŀ
    0.08
    Tex
    0.08
     Tex
    0.08
    人éĻħ
    0.08
    ä¼łçľŁ
    0.08
    çºŃ
    0.07
    enz
    0.07
    Activations Density 0.477%

    No Known Activations