© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 47576
    Prev
    Next
    INDEX
    Explanations

    states or conditions caused by

    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
    Negative Logits
    èĨº
    -0.10
    иков
    -0.10
    èĥ¸åīį
    -0.09
    волÑĮ
    -0.09
    оÑī
    -0.09
    æŃ¥è¡Į
    -0.09
     기íĥĢ
    -0.09
    å®Ŀè´µçļĦ
    -0.08
     плаÑģÑĤ
    -0.08
    ведение
    -0.08
    POSITIVE LOGITS
    éĴĬ
    0.09
    ãģįãģ£ãģĭãģij
    0.09
     reason
    0.09
    æĺ¯åĽłä¸º
    0.09
    Ä¡
    0.08
    ãĥ¼ãĥ³
    0.08
    æĺ¯ä¸ºäºĨ
    0.08
    åŃľ
    0.08
    /basic
    0.08
    IFS
    0.08
    Activations Density 0.139%

    No Known Activations