© 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. 62239
    Prev
    Next
    INDEX
    Explanations

    write, take, or use

    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
    iti
    -0.08
    rice
    -0.08
    æ¶ī
    -0.08
     diy
    -0.08
    æľĢæľī
    -0.08
    ijo
    -0.08
    èµIJ
    -0.08
    å¼ĢæĿ¥
    -0.08
     Cathy
    -0.08
    à¹Ģศรษà¸IJ
    -0.08
    POSITIVE LOGITS
    åĩıåĢ¼
    0.11
    iners
    0.10
    æįŁå¤±
    0.09
    计æıIJ
    0.09
    å¹´æľ«
    0.09
    人éĹ´
    0.09
    æĹ©æĹ©
    0.09
    æķijåĬ©
    0.09
    åı¸
    0.08
    åĮħ袱
    0.08
    Activations Density 0.021%

    No Known Activations