© 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. 20277
    Prev
    Next
    INDEX
    Explanations

    act of, action of

    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
    azo
    -0.09
     Lazar
    -0.08
    宵
    -0.08
    aza
    -0.08
    ä»ģ
    -0.08
    _SO
    -0.08
     unused
    -0.08
    åħ¶æ¬¡æĺ¯
    -0.07
    éºŁ
    -0.07
    æ¯į
    -0.07
    POSITIVE LOGITS
    è¿ĩç¨ĭ
    0.13
    è¿ĩç¨ĭä¸Ń
    0.12
    æģ°
    0.11
    éģİç¨ĭ
    0.11
    éģİç¨ĭä¸Ń
    0.11
    è¿ĩç¨ĭå½ĵä¸Ń
    0.10
    çļĦè¿ĩç¨ĭ
    0.10
    inan
    0.10
     process
    0.10
    çļĦè¿ĩç¨ĭä¸Ń
    0.09
    Activations Density 0.050%

    No Known Activations