© 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. 4588
    Prev
    Next
    INDEX
    Explanations

    turn on

    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
    çŀ¿
    -0.09
    ám
    -0.09
    buster
    -0.09
    invitation
    -0.09
     Dirk
    -0.09
    çijŁ
    -0.08
    itech
    -0.08
    orne
    -0.08
    hoe
    -0.08
    åĿļå®ļä¸į
    -0.08
    POSITIVE LOGITS
    ear
    0.09
    ä¸įæĺİçĻ½
    0.09
    æīī
    0.08
    å¤ĩçĶ¨
    0.08
    поÑĩ
    0.08
    å·²ç»ıæĪIJ为
    0.08
    æ°Ķæ°Ľ
    0.08
    /on
    0.08
     ÐŀÐŀ
    0.08
    mutable
    0.08
    Activations Density 0.016%

    No Known Activations