© 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. 4848
    Prev
    Next
    INDEX
    Explanations

    sanctions or removal

    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
    ä¹īåĬ¡
    -0.09
    eur
    -0.09
    urate
    -0.09
    orts
    -0.09
    èĥ½å¤Łåľ¨
    -0.09
    åŁķ
    -0.09
    éĤ£è¾¹
    -0.08
    åĿıäºĨ
    -0.08
    >Main
    -0.08
    destruct
    -0.08
    POSITIVE LOGITS
    åĪ°æľŁ
    0.09
     vapor
    0.09
    åĪ°ä½į
    0.08
     Bru
    0.08
    æī¾ä¸įåĪ°
    0.08
    åıĤåĬłä¼ļè®®
    0.08
    è·ĭ
    0.08
     Grove
    0.08
    å¦ĤæľŁ
    0.08
    ä¸ĢæĹ¥
    0.08
    Activations Density 0.007%

    No Known Activations