© 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. 34225
    Prev
    Next
    INDEX
    Explanations

    plenty of

    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¬ijçĿĢ说
    -0.09
    ick
    -0.09
    è¹¼
    -0.09
     вполне
    -0.09
    åľ£åľ°
    -0.09
    sWith
    -0.09
    赶快
    -0.08
    ìį¨
    -0.08
    ุ
    -0.08
    ç®Ģ缴æĺ¯
    -0.08
    POSITIVE LOGITS
    erno
    0.11
    iros
    0.10
    åħ¶ä»ĸçļĦ
    0.10
    borg
    0.10
    åħ¶ä»ĸ
    0.10
    jab
    0.09
     дÑĢÑĥгие
    0.09
    íŀĪ
    0.09
    jsc
    0.09
    çļĦä¸ĢéĿ¢
    0.09
    Activations Density 0.011%

    No Known Activations