© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52965
    Prev
    Next
    INDEX
    Explanations

    auto-generated content

    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
    ç®Ģ缴
    -0.10
     Pey
    -0.09
    æ¸ħæ¥ļ
    -0.09
    elman
    -0.09
    .slim
    -0.09
    æ¢Ĩ
    -0.09
    ç®Ģ缴æĺ¯
    -0.09
    èĽĬ
    -0.09
     wasted
    -0.09
    isman
    -0.09
    POSITIVE LOGITS
     automated
    0.11
    èĩªåĬ¨åĮĸ
    0.09
     Stafford
    0.09
    èĩªåĬ¨
    0.09
    agi
    0.09
     Alg
    0.09
    owo
    0.08
    åĴĮ个人
    0.08
     automatic
    0.08
     automatically
    0.08
    Activations Density 0.055%

    No Known Activations