© 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. 3237
    Prev
    Next
    INDEX
    Explanations

    Red Hat

    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
    UAL
    -0.09
    æĮª
    -0.08
    /operators
    -0.08
    -routing
    -0.08
    ively
    -0.08
    æĥ°
    -0.08
    릿
    -0.08
    orer
    -0.08
    ly
    -0.07
    _router
    -0.07
    POSITIVE LOGITS
    éļĶéŁ³
    0.10
    иÑģк
    0.10
    æĻ¡
    0.09
    ogle
    0.09
    askan
    0.09
    ainen
    0.09
    çĦIJ
    0.09
    ä¸įçķĻ
    0.08
    å§Ķ书记
    0.08
    inct
    0.08
    Activations Density 0.015%

    No Known Activations