© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10510
    Prev
    Next
    INDEX
    Explanations

    light, day, time, conditions

    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
    æµģåĬ¨
    -0.09
    èĮ¶åĽŃ
    -0.09
    éĤ£åıª
    -0.09
     Toe
    -0.08
    管
    -0.08
    ä¸ĸ
    -0.08
     пен
    -0.08
     overflowing
    -0.08
    è®¤çľŁåŃ¦ä¹ł
    -0.08
    容
    -0.08
    POSITIVE LOGITS
    åħī线
    0.11
    æĺ¼å¤ľ
    0.11
    åħī
    0.11
    åħīçħ§
    0.10
    mast
    0.10
    lit
    0.09
    yat
    0.09
    æŀŃ
    0.09
    yal
    0.09
     Edwards
    0.09
    Activations Density 0.090%

    No Known Activations