© 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. 57315
    Prev
    Next
    INDEX
    Explanations

    too X to be Y

    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
    бл
    -0.10
     ÑħозÑıй
    -0.10
    PCS
    -0.09
    olem
    -0.09
    æĪij们åºĶ该
    -0.09
    æľīå¤ļä¹Ī
    -0.09
    éĽĨæĪIJçĶµè·¯
    -0.09
    å°¤åħ¶æĺ¯
    -0.09
     Chill
    -0.08
    _objects
    -0.08
    POSITIVE LOGITS
    åĽ½åĬ¡
    0.10
    害
    0.09
    вин
    0.08
    è°Ĺ
    0.08
     cherry
    0.08
    æľīçļĦ人
    0.08
    ç°¸
    0.08
     ÙĪغÙĬر
    0.08
    hear
    0.08
    èĬ±éĴ±
    0.08
    Activations Density 0.003%

    No Known Activations