© 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. 2730
    Prev
    Next
    INDEX
    Explanations

    special characters or deep

    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
    VIC
    -0.09
    åĮ£
    -0.09
    該使çĶ¨èĢħ
    -0.09
    ä»»ä½ķ人
    -0.09
    åľ°éģĵ
    -0.09
    ãĥ¡ãĥĩãĤ£
    -0.08
    禧
    -0.08
    çIJ¼
    -0.08
    yre
    -0.08
    .TestCase
    -0.08
    POSITIVE LOGITS
    çĻ¾äº¿
    0.10
    太éĺ³
    0.10
    åĿł
    0.09
    éľĩæĴ¼
    0.09
    å®ľå±ħ
    0.09
    çĮİ
    0.09
     deep
    0.09
    åįĥ亿
    0.09
    çĤ½
    0.08
     Deep
    0.08
    Activations Density 0.201%

    No Known Activations