© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9966
    Prev
    Next
    INDEX
    Explanations

    moving into or onto

    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
     Gör
    -0.09
    å³»
    -0.09
    burg
    -0.09
    鼳
    -0.09
    åħ¥éŨ
    -0.09
     cul
    -0.09
    æĨ©
    -0.09
    erner
    -0.08
    tte
    -0.08
    ras
    -0.08
    POSITIVE LOGITS
    èĮ«èĮ«
    0.12
    ä¸ĸ人
    0.12
    å¤ĸçķĮ
    0.11
    åħ¨ç¤¾ä¼ļ
    0.10
     окÑĢÑĥжа
    0.10
    çİ°å®ŀä¸Ń
    0.10
    åħ¬ä¼Ĺ
    0.10
    é£İæ°Ķ
    0.10
    AIN
    0.10
    å¤ĸéĿ¢
    0.10
    Activations Density 0.057%

    No Known Activations