© 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. 20026
    Prev
    Next
    INDEX
    Explanations

    make/made verbs

    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
    èģĮ称
    -0.10
    æĺŁçIJĥ
    -0.09
    æĢĢ念
    -0.09
    éĴ±è´¢
    -0.09
     Inch
    -0.09
     деле
    -0.09
    æĥ³èµ·æĿ¥
    -0.09
    .mutable
    -0.08
    ä¸ĢçĤ¹çĤ¹
    -0.08
     Olsen
    -0.08
    POSITIVE LOGITS
    强迫
    0.12
    强åĪ¶
    0.11
    éĢ¼
    0.10
    æĭīçĿĢ
    0.10
    éªĬ
    0.10
    ANGER
    0.09
    太æŀģ
    0.09
    ä¸įåĩĨ
    0.08
    èĭĽ
    0.08
    è¦ģæ±Ĥ
    0.08
    Activations Density 0.039%

    No Known Activations