© 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. 27053
    Prev
    Next
    INDEX
    Explanations

    consecutive wins/losses in games

    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
    çĭĻ
    -0.10
    çŃ·åŃIJ
    -0.09
    åķ¾
    -0.09
    oret
    -0.09
    Intialized
    -0.09
    éķĸ
    -0.08
     excer
    -0.08
    (Android
    -0.08
    为ä¼ģä¸ļ
    -0.08
    éĶ·
    -0.08
    POSITIVE LOGITS
    äººåľ¨
    0.10
    åIJĮæĹ¶ä¹Łæĺ¯
    0.10
    æķħ
    0.09
    è¯ķ
    0.09
    _masks
    0.09
     attempt
    0.08
    å°Ŀè¯ķ
    0.08
     Progressive
    0.08
    éĩįè¦ģ
    0.08
    éĩį大
    0.08
    Activations Density 0.037%

    No Known Activations