© 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. 33087
    Prev
    Next
    INDEX
    Explanations

    sure when

    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
    æľĶ
    -0.09
    ç®Ģ缴æĺ¯
    -0.08
    çĵ¶é¢Ī
    -0.08
    åıªæĥ³
    -0.08
     utmost
    -0.08
    æľ½
    -0.08
    åıªèĥ½è¯´
    -0.08
    è¯ķ管
    -0.07
    argo
    -0.07
    soup
    -0.07
    POSITIVE LOGITS
    endon
    0.10
     Vend
    0.10
    æĸ°é²ľ
    0.10
    acket
    0.09
    thon
    0.09
    sto
    0.09
    beck
    0.09
    åĪ¤
    0.09
    .Editor
    0.09
    æĸ¹ä¾¿
    0.09
    Activations Density 0.009%

    No Known Activations