© 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. 58281
    Prev
    Next
    INDEX
    Explanations

    any

    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
     sometimes
    -0.09
     иногда
    -0.09
    æľīæŶ
    -0.09
    ãģĤãĤĭ
    -0.09
    æŁIJäºĽ
    -0.09
    èĢĮä¸įæĺ¯
    -0.09
    ä¸Ģå®ļçļĦ
    -0.09
     certain
    -0.09
    æľīæŶåĢĻ
    -0.09
    æ¯ı个人çļĦ
    -0.08
    POSITIVE LOGITS
     whatsoever
    0.19
     WHATSOEVER
    0.13
    åıªè¦ģæĺ¯
    0.13
    éĥ½åı¯ä»¥
    0.12
    åĵªæĢķæĺ¯
    0.12
    THING
    0.12
    åŁºæľ¬ä¸Ĭéĥ½
    0.12
    /all
    0.11
    (any
    0.11
     سÙĪÙī
    0.11
    Activations Density 0.566%

    No Known Activations