© 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. 27702
    Prev
    Next
    INDEX
    Explanations

    star names and roles

    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
    侦
    -0.09
    æľīåħ´è¶£
    -0.08
    tod
    -0.08
    éĿĸ
    -0.08
    åIJĦä½į
    -0.08
    ä»ĸ说
    -0.08
    \\/
    -0.08
    姹
    -0.08
    å¼ĥ
    -0.08
    éĵł
    -0.08
    POSITIVE LOGITS
    asting
    0.10
    мон
    0.10
    .itemView
    0.10
    URRE
    0.10
    æīĢåľ¨
    0.10
    LING
    0.09
     cán
    0.09
     Nets
    0.09
    _GRANTED
    0.09
    éĩįçİ°
    0.09
    Activations Density 0.012%

    No Known Activations