© 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. 22999
    Prev
    Next
    INDEX
    Explanations

    unannounced surprise warning unaware

    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
    ØŃÙĦ
    -0.09
    LAY
    -0.09
    åĴĢ
    -0.09
    setProperty
    -0.09
    éģĹæĨ¾
    -0.08
    å¡«
    -0.08
    aurus
    -0.08
    æĽ¿
    -0.08
     Jaw
    -0.08
    è¯Ń
    -0.08
    POSITIVE LOGITS
    ä¸įå®ļ
    0.12
    é¢Ħ约
    0.11
    æĿ¥è®¿
    0.11
    çİĦ
    0.11
    vas
    0.11
    çĮĿ
    0.11
    çªģåĩ»
    0.11
    äºĭåħĪ
    0.11
    VR
    0.10
    æĬ½æŁ¥
    0.10
    Activations Density 0.070%

    No Known Activations