© 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. 23159
    Prev
    Next
    INDEX
    Explanations

    increase or decrease

    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
    icap
    -0.10
    IAL
    -0.10
    stalk
    -0.09
    èĴĻ
    -0.09
    bao
    -0.09
     подпиÑģ
    -0.09
    endar
    -0.09
    icom
    -0.09
    /native
    -0.08
    ühr
    -0.08
    POSITIVE LOGITS
    è¿ĻäºĽå¹´
    0.14
     trends
    0.12
     Trends
    0.12
    è¶ĭåĬ¿
    0.11
    çļĦè¶ĭåĬ¿
    0.11
    èĩªä»İ
    0.10
    çļĦåıĺåĮĸ
    0.10
    ä¿®åħ»
    0.10
    è¶ĬæĿ¥è¶Ĭ
    0.10
    è¿Ļä¹Īå¤ļå¹´
    0.10
    Activations Density 0.329%

    No Known Activations