© 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. 53560
    Prev
    Next
    INDEX
    Explanations

    prediction accuracy

    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
    okin
    -0.09
    åıĸè¯ģ
    -0.09
    çªģåĽ´
    -0.09
    æıIJè®®
    -0.09
    èĥŀ
    -0.08
    ä½ĵä¼ļåĪ°
    -0.08
    åĿĤ
    -0.08
    峪
    -0.08
     Hell
    -0.08
     Ryu
    -0.08
    POSITIVE LOGITS
    é¢Ħæµĭ
    0.15
    åĩĨç¡®æĢ§
    0.14
    é¢ĦæĬ¥
    0.14
    å¯¹æľªæĿ¥
    0.13
    åĩĨç¡®
    0.11
     accuracy
    0.11
    accuracy
    0.11
     Accuracy
    0.11
    Accuracy
    0.11
    å±ķæľĽ
    0.11
    Activations Density 0.138%

    No Known Activations