© 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. 60995
    Prev
    Next
    INDEX
    Explanations

    following or given

    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
    åīįæıIJæĺ¯
    -0.10
    èIJ¼
    -0.09
    åĢ¦
    -0.09
    retch
    -0.09
    ndon
    -0.09
    ick
    -0.09
    ä¸įçĦ¶
    -0.09
    witter
    -0.09
    æĬ¢
    -0.09
    кÑĢ
    -0.09
    POSITIVE LOGITS
    å¦Ĥä¸ĭ
    0.15
     below
    0.15
     ниже
    0.14
    åľ¨è¿ĻéĩĮ
    0.13
     here
    0.13
    ç»Ļä½łä»¬
    0.12
    below
    0.11
    è¿ĻéĩĮ
    0.11
    ä¸ĭéĿ¢å°±
    0.11
    åĪĹåĩº
    0.11
    Activations Density 0.239%

    No Known Activations