© 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. 57057
    Prev
    Next
    INDEX
    Explanations

    long-term follow-up, retrospective studies

    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
    å¯ĨéĽĨ
    -0.10
    oley
    -0.09
    èľ»
    -0.09
     вÑģÑĤÑĢ
    -0.08
    .MAIN
    -0.08
    ampo
    -0.08
    ç°ª
    -0.08
    深深
    -0.08
    彩
    -0.08
    åīįä¸ī
    -0.08
    POSITIVE LOGITS
    æ°¸ä¹ħ
    0.15
    ç»Ī身
    0.12
    继ç»Ń
    0.11
    åĩłå¹´
    0.11
    ä¸Ģå¹´
    0.11
     permanently
    0.10
     permanent
    0.10
    ä¸ĥå¹´
    0.10
    æ°¸è¿ľ
    0.10
     forever
    0.10
    Activations Density 0.718%

    No Known Activations