© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10547
    Prev
    Next
    INDEX
    Explanations

    what we're doing

    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
    ä¸įåIJĮäºİ
    -0.10
     zwar
    -0.09
    ä¸įä»ħæĺ¯
    -0.09
     both
    -0.09
     however
    -0.08
    è¿ĺæĺ¯ä¼ļ
    -0.08
    大æ¦Ĥ
    -0.08
    _rw
    -0.08
    éĿŀ常
    -0.08
    çĶļèĩ³è¿ĺ
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.15
    èĢĮæĺ¯
    0.13
    åįķ纯çļĦ
    0.13
     ones
    0.11
    æģ°æģ°
    0.11
    éĤ£ä¹Ī容æĺĵ
    0.11
    ÙĤÙħ
    0.11
     necessarily
    0.10
     ÑģобÑģÑĤвенно
    0.10
    åįķåįķ
    0.10
    Activations Density 0.169%

    No Known Activations