© 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. 58849
    Prev
    Next
    INDEX
    Explanations

    sports actions and plays

    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
    èĻIJ
    -0.09
    å¹¼
    -0.09
    ember
    -0.09
    éļIJ约
    -0.08
     remin
    -0.08
    oster
    -0.08
    lico
    -0.08
    ibel
    -0.08
     нек
    -0.08
    æīĢå±ŀ
    -0.08
    POSITIVE LOGITS
     Advantage
    0.10
    ä¸ŃåĽ½éĺŁ
    0.09
    失误
    0.09
    让èĩªå·±
    0.09
    红åĪ©
    0.09
    大åħ¨
    0.09
    大åĨĽ
    0.09
    ((&
    0.09
     вÑĭгод
    0.09
    æĢĢçĸij
    0.09
    Activations Density 0.010%

    No Known Activations