© 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. 31285
    Prev
    Next
    INDEX
    Explanations

    replaced by something else

    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
    çĩİ
    -0.12
    åĩĢ
    -0.10
    æ·¨
    -0.09
    Arthur
    -0.08
    èĮ¸
    -0.08
    椽
    -0.08
    _mB
    -0.08
    è´¨åľ°
    -0.08
    hma
    -0.08
    .CODE
    -0.08
    POSITIVE LOGITS
    æį¢æĪIJ
    0.35
     replaced
    0.28
    æĶ¹ä¸º
    0.23
    æĽ¿æį¢
    0.22
    æĶ¹æĪIJ
    0.22
     replace
    0.22
    转åıĺ为
    0.21
     Replace
    0.19
    åıĺæĪIJäºĨ
    0.19
    åıĺ为
    0.19
    Activations Density 0.210%

    No Known Activations