© 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. 3233
    Prev
    Next
    INDEX
    Explanations

    wood and deciduous trees

    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
    weise
    -0.10
    arra
    -0.09
     Gallup
    -0.08
    aurus
    -0.08
    reh
    -0.08
    oen
    -0.08
    ä¼İ
    -0.08
    witter
    -0.08
    arah
    -0.07
     поÑģледние
    -0.07
    POSITIVE LOGITS
    èģĶèµĽ
    0.11
    æ¯ĭ
    0.09
     floors
    0.09
    åľĪ
    0.09
    utes
    0.09
    çļij
    0.09
    .Lang
    0.09
    ç»ĻäºĨæĪij
    0.08
     Bridges
    0.08
     sÃłng
    0.08
    Activations Density 0.002%

    No Known Activations