© 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. 57432
    Prev
    Next
    INDEX
    Explanations

    excel or excels verb

    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
    åŃ¦å®¶
    -0.09
    acks
    -0.09
    cip
    -0.09
    /people
    -0.08
    s
    -0.08
    3
    -0.08
    icus
    -0.08
    ties
    -0.08
     seriously
    -0.08
    about
    -0.08
    POSITIVE LOGITS
    erator
    0.13
    ä»»ä½ķæŶåĢĻ
    0.12
    ä¹ĭå¤Ħ
    0.11
    erate
    0.11
    åħ¶éĹ´
    0.10
    ible
    0.10
    æĭįæĭį
    0.10
    UME
    0.10
    REW
    0.10
    ential
    0.10
    Activations Density 0.012%

    No Known Activations