© 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. 32448
    Prev
    Next
    INDEX
    Explanations

    for our, for your, for both

    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
    deen
    -0.09
    深深çļĦ
    -0.09
    åı£è¢ĭ
    -0.09
    é»ijå¤ľ
    -0.09
    渲
    -0.09
    çļĦ巨大
    -0.08
    大èĩªçĦ¶
    -0.08
    åı£ä¸Ń
    -0.08
    FER
    -0.08
    å¾·æĭī
    -0.08
    POSITIVE LOGITS
     bè
    0.12
    ä»į
    0.10
    çĬ¹
    0.09
    çĭ¬
    0.09
    ä¸ĭ次
    0.09
    /latest
    0.09
    _ACCEPT
    0.09
    /dat
    0.09
    çĭ¬èĩª
    0.09
    adopt
    0.08
    Activations Density 0.065%

    No Known Activations