© 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. 5637
    Prev
    Next
    INDEX
    Explanations

    now

    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
    æѤæŶ
    -0.10
    orce
    -0.10
    ä»İæѤ
    -0.09
    ãģĦãģ¾
    -0.09
    loth
    -0.08
    ä»ĺ
    -0.08
    airport
    -0.08
    æѤæĻĤ
    -0.08
    çĢijå¸ĥ
    -0.08
    edImage
    -0.08
    POSITIVE LOGITS
    adays
    0.12
    åĽŀæĥ³
    0.11
    å·²æĺ¯
    0.10
    å¢ŀèĩ³
    0.10
     пÑĢинÑı
    0.09
    è¶ħè¿ĩ
    0.09
    ise
    0.09
    被称为
    0.09
    èĽĨ
    0.09
    æĬ¥éģĵ称
    0.09
    Activations Density 0.036%

    No Known Activations