© 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. 34406
    Prev
    Next
    INDEX
    Explanations

    upgrade to

    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
    äºĶåĪĨéĴŁ
    -0.11
    ä¸ĢåĪĨéĴŁ
    -0.11
    åĩłåĪĨéĴŁ
    -0.09
    å·¥åľ°
    -0.09
    æľīçĶ¨çļĦ
    -0.09
    /ns
    -0.08
    æ¯ĶèµĽ
    -0.08
    à¸ķรา
    -0.08
    åı¯éĿłçļĦ
    -0.08
    æıŃå¼Ģ
    -0.08
    POSITIVE LOGITS
    åįĩ级
    0.19
     upgrade
    0.16
     upgrading
    0.16
     Upgrade
    0.15
    Upgrade
    0.14
    upgrade
    0.14
     upgrades
    0.13
    æŧ
    0.12
     upgraded
    0.12
    _upgrade
    0.12
    Activations Density 0.112%

    No Known Activations