© 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. 57879
    Prev
    Next
    INDEX
    Explanations

    writes:"

    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
    anche
    -0.08
    ÑĤол
    -0.08
    cone
    -0.08
    åīĮ
    -0.08
    çĬĩ
    -0.08
    nonce
    -0.08
    ש×ķ×ij
    -0.08
    uper
    -0.08
    ei
    -0.08
    .INSTANCE
    -0.08
    POSITIVE LOGITS
    çļĦæľĢåIJİä¸Ģ
    0.10
    éģĴ
    0.09
    ä¸Ģåı¥è¯Ŀ
    0.09
    ä¸ĢåĪĩéĥ½
    0.09
    ä¸Ģ人
    0.09
    raph
    0.08
    ä¸Ģåı¥
    0.08
    Caught
    0.08
    ä¸Ģå¹´ä¸Ģ度
    0.08
    è¿ĩåİ»çļĦ
    0.08
    Activations Density 0.018%

    No Known Activations