© 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. 35467
    Prev
    Next
    INDEX
    Explanations

    ball sports

    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
    导
    -0.11
     Fried
    -0.10
    èĤĺ
    -0.10
    é³Ŀ
    -0.10
    åĨ·åį´
    -0.09
    èħ°
    -0.09
    ä½ķå¿ħ
    -0.09
    ä¼ĺåĬ¿
    -0.09
    VERS
    -0.09
    å°İ
    -0.09
    POSITIVE LOGITS
    çIJĥ
    0.13
    ãĥľãĥ¼ãĥ«
    0.12
     ball
    0.11
    (ball
    0.10
    æ¡´
    0.10
    ball
    0.10
     balls
    0.09
    纸质
    0.09
    balls
    0.09
    _sphere
    0.09
    Activations Density 0.072%

    No Known Activations