© 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. 4690
    Prev
    Next
    INDEX
    Explanations

    preceding or following punctuation

    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
    nex
    -0.08
    yst
    -0.08
    кÑĥÑĢ
    -0.08
    assing
    -0.08
     gamble
    -0.08
    olas
    -0.07
    å®Ĺ
    -0.07
    å°¤åħ¶æĺ¯
    -0.07
    ène
    -0.07
    Penn
    -0.07
    POSITIVE LOGITS
    æĦıæĢĿæĺ¯
    0.12
    ä»Ģä¹ĪæĦıæĢĿ
    0.10
    ä¸ĢåĽŀäºĭ
    0.10
    losures
    0.09
    ä¿Ĺ称
    0.09
    æĦıæĢĿ
    0.09
    æ³Ħéľ²
    0.09
    æľ½
    0.09
    çłĶ究ä¸Ńå¿ĥ
    0.08
    illon
    0.08
    Activations Density 0.331%

    No Known Activations