© 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. 32835
    Prev
    Next
    INDEX
    Explanations

    said, saw, was, never

    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
    èĥ½è®©
    -0.10
    ä¼ĹæīĢ
    -0.10
    çľ¼ä¸ĭ
    -0.10
    ç±į
    -0.09
     presently
    -0.09
    è¿Ħä»Ĭ
    -0.09
     apprent
    -0.09
    è¿Ħä»Ĭ为æŃ¢
    -0.09
    ä¸Ģ个æĸ°çļĦ
    -0.09
    å¾Ĵ
    -0.09
    POSITIVE LOGITS
    æĪijå½ĵæŶ
    0.11
    å½ĵæŶ
    0.11
    åĨĹ
    0.10
    åijĬçŁ¥
    0.10
    å¿ĥéĩĮ
    0.10
    å¿ĥæĥ³
    0.09
    å°±è§īå¾Ĺ
    0.09
    å½ĵæŶçļĦ
    0.09
    çķ¶æĻĤ
    0.08
     подгоÑĤов
    0.08
    Activations Density 0.177%

    No Known Activations