© 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. 34504
    Prev
    Next
    INDEX
    Explanations

    change or reversal

    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
    ige
    -0.11
    å¿Ĺ
    -0.10
    å§ĵ
    -0.10
    ä¸įåIJĮç¨ĭ度
    -0.09
    æĸ°çīĪ
    -0.09
    æįº
    -0.09
    èĪŀåı°
    -0.09
    ì¡°ìĤ¬
    -0.09
    ä¸įä¸ĭ
    -0.08
    èĭĶ
    -0.08
    POSITIVE LOGITS
    é¢ĵ
    0.11
    çĩĥ
    0.10
    UES
    0.10
    æ¸ħæĺİ
    0.09
     ÑģиÑĤÑĥаÑĨии
    0.09
    -Feb
    0.09
    å±ĢéĿ¢
    0.09
     situación
    0.08
    enco
    0.08
    çļĦåij½è¿IJ
    0.08
    Activations Density 0.020%

    No Known Activations