© 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. 30089
    Prev
    Next
    INDEX
    Explanations

    key to success

    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
     воÑĤ
    -0.10
    ç«ŀäºī对æīĭ
    -0.08
     канал
    -0.08
    æĺ¯ä»¥
    -0.08
    ix
    -0.08
    æ´µ
    -0.08
    èĨı
    -0.08
    åĬ«
    -0.08
     Bek
    -0.08
    orth
    -0.07
    POSITIVE LOGITS
     understanding
    0.12
     success
    0.12
    success
    0.11
    æķ´ä¸ªäºº
    0.10
    æŃ£å¸¸ä½¿çĶ¨
    0.10
    çIJĨ解
    0.09
    çļĦæĪIJåĬŁ
    0.09
    社ä¼ļ稳å®ļ
    0.09
    .success
    0.09
     Success
    0.09
    Activations Density 0.053%

    No Known Activations