© 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. 54476
    Prev
    Next
    INDEX
    Explanations

    service members and responders

    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
    褥
    -0.10
    åĬ¨æīĭ
    -0.09
    ä¸Ģ个å°ı
    -0.08
     punct
    -0.08
    åīIJ
    -0.08
    åIJĪçIJĨçļĦ
    -0.08
    ä¸Ģ棵
    -0.08
    åĪĨæīĭ
    -0.08
     exp
    -0.08
    losing
    -0.08
    POSITIVE LOGITS
    çīºçī²
    0.12
    onta
    0.10
    èĩ´æķ¬
    0.10
    çĬ
    0.10
    åĽ½å®¶å®īåħ¨
    0.10
    å®īåħ¨ä¿Ŀéļľ
    0.09
    å½ĵ代
    0.09
    éĥµ
    0.09
    å¿ħä¸įåı¯
    0.09
    vic
    0.09
    Activations Density 0.028%

    No Known Activations