© 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. 19875
    Prev
    Next
    INDEX
    Explanations

    written content

    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
    éģ¥
    -0.09
    çģŃ
    -0.09
    ÑĢÑİ
    -0.09
    è´¨
    -0.08
    箫
    -0.08
    ãĥ«ãĥī
    -0.08
    ucked
    -0.08
     ÑĢавно
    -0.08
    /cop
    -0.08
    åij½ä»¤
    -0.08
    POSITIVE LOGITS
    书éĿ¢
    0.10
    课åłĤæķĻåѦ
    0.09
    åIJ¬çĿĢ
    0.09
    éĽī
    0.09
     Donovan
    0.09
    课åłĤ
    0.08
    addListener
    0.08
    listening
    0.08
    åIJ¬äºĨ
    0.08
    å½ĵåľº
    0.08
    Activations Density 0.286%

    No Known Activations