© 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. 58629
    Prev
    Next
    INDEX
    Explanations

    over and over again

    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¬ī
    -0.09
    çľ¼æ³ª
    -0.09
    uxe
    -0.09
    æľīä¸Ģ天
    -0.08
    æ°¸ä¹ħ
    -0.08
    à¸ģาร
    -0.08
    ä¸įä¸ĭ
    -0.08
    mobx
    -0.08
    羣çļĦæĺ¯
    -0.08
    itical
    -0.08
    POSITIVE LOGITS
    -place
    0.11
    è¹Ī
    0.10
    赢家
    0.09
    ä½ķ度ãĤĤ
    0.09
    ä¸įçĦ¶
    0.09
    åıįå¤į
    0.09
    çĽ¸ä¼ł
    0.09
    èĢĥéªĮ
    0.09
    ä¸Ģéģį
    0.08
    ently
    0.08
    Activations Density 0.007%

    No Known Activations