© 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. 56200
    Prev
    Next
    INDEX
    Explanations

    engine parts and mechanics

    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
    æ´ģçĻ½
    -0.08
    Bas
    -0.08
    åĬ²
    -0.08
    ãģ¨ãĤĤ
    -0.08
    вой
    -0.08
    ç±³ç²ī
    -0.08
    æѤåĪ»
    -0.08
    åĭģ
    -0.08
    ieber
    -0.08
    bas
    -0.08
    POSITIVE LOGITS
    éĿ¢åŃĶ
    0.09
    主åĬŀ
    0.09
    inqu
    0.09
    åĪ¤
    0.08
    icator
    0.08
    èĬĤå¥ı
    0.08
    åºŀ
    0.08
    è¶´åľ¨
    0.08
    宽容
    0.08
    æİĮ
    0.08
    Activations Density 0.032%

    No Known Activations