© 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. 33085
    Prev
    Next
    INDEX
    Explanations

    “We " or “If "

    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¸ĩæĪ·
    -0.08
    4
    -0.08
    -spin
    -0.08
     rit
    -0.08
    寰
    -0.08
     ray
    -0.08
    7
    -0.08
     inform
    -0.08
    æĦģ
    -0.08
    åѦçĶŁä»¬
    -0.08
    POSITIVE LOGITS
    ãģĿãĤĮãģ¯
    0.09
    åŁİéĻħ
    0.08
    resher
    0.08
    vement
    0.08
    æľīä¸Ģ次
    0.08
    \Backend
    0.08
    åģĮ
    0.08
    ÑĢег
    0.08
    è°ģèĥ½
    0.08
    æ¶Īæŀģ
    0.08
    Activations Density 0.029%

    No Known Activations