© 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. 3343
    Prev
    Next
    INDEX
    Explanations

    zandvoort, chaco, sungai, kidanganad, otolaryngology, knee, aldermen, gulf, elicitins, marina

    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
    æŁ¬
    -0.10
    odore
    -0.10
    éľĸ
    -0.09
    æłĭ
    -0.09
    ussian
    -0.09
    burg
    -0.09
    ç»ıçIJĨ
    -0.09
    太太
    -0.09
    OrNull
    -0.09
    ç¯ĵ
    -0.09
    POSITIVE LOGITS
    èĢģ大
    0.08
    åĬ³
    0.08
    ecz
    0.08
     Bas
    0.07
    ä¸īæĸ¹
    0.07
    äºijéĽĨ
    0.07
    åħ¶æīĢ
    0.07
    zza
    0.07
    ä¼Ĭæĭī
    0.07
    ä¼ij
    0.07
    Activations Density 0.314%

    No Known Activations