© 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. 29906
    Prev
    Next
    INDEX
    Explanations

    normal or usual

    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
    reira
    -0.09
    éĹªéĹª
    -0.09
    寡
    -0.09
    å½ĵåį³
    -0.09
    澡
    -0.08
    乡
    -0.08
    åĴĴ
    -0.08
    /license
    -0.08
    åĬ¡å¿ħ
    -0.08
    ละ
    -0.08
    POSITIVE LOGITS
    æĻ®éĢļçļĦ
    0.16
    æŃ£å¸¸çļĦ
    0.16
    æĻ®éĢļ
    0.16
    常è§Ħ
    0.13
     обÑĭÑĩ
    0.13
    ä¼łç»ŁçļĦ
    0.12
    /plain
    0.12
    normal
    0.12
     ordinary
    0.12
    æŃ£å¸¸
    0.11
    Activations Density 0.241%

    No Known Activations