© 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. 57607
    Prev
    Next
    INDEX
    Explanations

    black and white images

    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
    chez
    -0.09
    è¢į
    -0.09
    æľºåĬ¨
    -0.09
    ively
    -0.09
    æĬµ
    -0.08
    æĹ¥åĨħ
    -0.08
    lse
    -0.08
    ÑģÑĤÑĥп
    -0.08
    èģĶåIJĪ
    -0.08
    dez
    -0.08
    POSITIVE LOGITS
    é»ijçĻ½
    0.17
    线æĿ¡
    0.11
    çģ°
    0.10
    çģ°èī²
    0.10
    å¹½
    0.10
    -ton
    0.09
    Forg
    0.09
    è¯ķçĤ¹å·¥ä½ľ
    0.09
    åºĸ
    0.09
    жÑĥ
    0.09
    Activations Density 0.050%

    No Known Activations