© 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. 2386
    Prev
    Next
    INDEX
    Explanations

    long-term goals or objectives

    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
    æĬ«
    -0.11
    yii
    -0.09
    æĢİä¹Ī说
    -0.08
    pl
    -0.08
    çıŀ
    -0.08
    ppo
    -0.08
    ä¹ĭ声
    -0.08
    é¾Ľ
    -0.08
    ä¿¡æģ¯åıijå¸ĥ
    -0.08
     Mun
    -0.07
    POSITIVE LOGITS
    缮æłĩ
    0.31
     goals
    0.29
    缮æ¨Ļ
    0.29
    çļĦ缮æłĩ
    0.27
     goal
    0.26
     objectives
    0.25
     mục
    0.25
    缮çļĦ
    0.25
    _goal
    0.23
     Goals
    0.23
    Activations Density 0.334%

    No Known Activations