© 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. 49348
    Prev
    Next
    INDEX
    Explanations

    punching or perforating

    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
    chant
    -0.10
    odash
    -0.09
    åį·
    -0.09
    è¢ħ
    -0.09
    åįģéĩĮ
    -0.09
    /is
    -0.08
    anter
    -0.08
    _maximum
    -0.08
    ä¸Ńéĥ¨
    -0.08
    \Notifications
    -0.08
    POSITIVE LOGITS
    holes
    0.11
    ../../../
    0.11
    å«¡
    0.09
    etti
    0.09
    /rfc
    0.09
    çĥ¯
    0.09
    (es
    0.08
    ../
    0.08
    ercise
    0.08
     Airways
    0.08
    Activations Density 0.011%

    No Known Activations