© 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. 48518
    Prev
    Next
    INDEX
    Explanations

    done, result, changing, good

    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
    åĺĽ
    -0.10
    èµĦæĸĻæĺ¾ç¤º
    -0.10
    osite
    -0.10
    ä»»ä½ķå½¢å¼ı
    -0.10
    æĹ¥æĬ¥ç¤¾
    -0.09
     Cheney
    -0.09
    ä¹Łä¸įä¾ĭå¤ĸ
    -0.09
    ä¸Ģå°ı
    -0.09
    รม
    -0.09
    rost
    -0.09
    POSITIVE LOGITS
    2
    0.12
    4
    0.12
    5
    0.12
    3
    0.11
    7
    0.11
    以åıĬ
    0.11
    6
    0.11
    9
    0.11
    8
    0.11
    1
    0.10
    Activations Density 0.512%

    No Known Activations