© 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. 33529
    Prev
    Next
    INDEX
    Explanations

    preset or specified conditions

    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
    éĤ£ä¹Īå¤ļ
    -0.10
    .gnu
    -0.10
    éģĹæ¼ı
    -0.09
    à¸ģà¸Ķ
    -0.09
    éĥ¨
    -0.09
    ÙĬتÙĩ
    -0.09
    æľīä»»ä½ķ
    -0.09
     Horn
    -0.09
    æĹ¥æĬ¥ç¤¾
    -0.09
    irts
    -0.09
    POSITIVE LOGITS
    æĮĩå®ļ
    0.14
     specified
    0.13
    è§Ħå®ļçļĦ
    0.12
     chosen
    0.12
    specified
    0.10
    çķĻç»Ļ
    0.10
     supplied
    0.10
    åĽºå®ļçļĦ
    0.10
    chosen
    0.10
     selected
    0.10
    Activations Density 0.460%

    No Known Activations