© 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. 31845
    Prev
    Next
    INDEX
    Explanations

    stories about

    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
     dot
    -0.08
    åį¡æĭī
    -0.08
    jer
    -0.08
     sort
    -0.08
    ä¸ĩè¾¾
    -0.08
    atum
    -0.08
     bada
    -0.08
    奥
    -0.07
     Tamb
    -0.07
    è¿ĺä¸į
    -0.07
    POSITIVE LOGITS
    æĬ¥éģĵ称
    0.12
    æĬ¥éģĵ
    0.12
    åł±å°İ
    0.12
    缸åħ³æĸ°éĹ»
    0.10
    æĬ¥è®°èĢħ
    0.10
    涸
    0.10
    èĪIJ
    0.09
    æŶéļĶ
    0.09
    äºī缸
    0.09
    æĻ¨æĬ¥
    0.09
    Activations Density 0.091%

    No Known Activations