© 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. 32146
    Prev
    Next
    INDEX
    Explanations

    genius, masterpiece, minds

    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¸įå°ij人
    -0.09
    ainen
    -0.09
    åı¥è¯Ŀ
    -0.09
    æ¯ħçĦ¶
    -0.09
    ToWorld
    -0.09
    jen
    -0.08
    hands
    -0.08
    å°¤åħ¶æĺ¯
    -0.08
    åij
    -0.08
     Baths
    -0.08
    POSITIVE LOGITS
    ous
    0.10
    ously
    0.10
    oft
    0.10
    天æīį
    0.09
    fully
    0.09
    ä¹ĭä½ľ
    0.09
    outu
    0.09
    å¢ĥçķĮ
    0.08
    OfType
    0.08
    isk
    0.08
    Activations Density 0.010%

    No Known Activations