© 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. 39499
    Prev
    Next
    INDEX
    Explanations

    transition metal elements

    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
    æīĢå¾Ĺ
    -0.10
    ç½į
    -0.09
    æĭį
    -0.09
    éĺĬ
    -0.08
    å«Ĥ
    -0.08
    æī¿è®¤
    -0.08
    ç®Ń
    -0.08
    æ·±åĮĸæĶ¹éĿ©
    -0.08
    æĪIJ为ä¸ŃåĽ½
    -0.08
    ç£ķ
    -0.08
    POSITIVE LOGITS
    é«ĺä»·
    0.09
    æ²¹çĶ»
    0.09
    èħ¾é£ŀ
    0.09
    æ£ķ
    0.09
    åĢ¾åIJij
    0.09
    èī²
    0.09
     êµ
    0.09
    èIJ½ä¸ĭ
    0.08
    ç»Łè®¡
    0.08
    _DECL
    0.08
    Activations Density 0.249%

    No Known Activations