© 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. 53429
    Prev
    Next
    INDEX
    Explanations

    geometry processing

    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
    éļ¶
    -0.11
    ContentLoaded
    -0.10
    èį¡
    -0.10
    èİ«åIJįåħ¶å¦Ļ
    -0.10
    åĨĻåŃĹ
    -0.10
    è¡¥é½IJ
    -0.10
    æľįåĬ¡ä¸ļ
    -0.10
    ocide
    -0.09
    éĽį
    -0.09
    IRO
    -0.09
    POSITIVE LOGITS
    穹
    0.09
     Nur
    0.09
     graft
    0.08
    æĴij
    0.08
     Glover
    0.08
    æĮ½
    0.08
    æĬ
    0.08
    ds
    0.08
    _support
    0.08
     opposite
    0.08
    Activations Density 0.071%

    No Known Activations