© 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. 30387
    Prev
    Next
    INDEX
    Explanations

    users can perform actions

    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
    åı¯èĥ½åıijçĶŁ
    -0.11
    å¨ĺ
    -0.11
    çĮĽåľ°
    -0.10
    éĴº
    -0.09
    å®ļåIJij
    -0.09
    ä¸įèĤ¯
    -0.09
    çļĦæł·åŃIJ
    -0.08
    æĪĸå°Ĩ
    -0.08
    主管
    -0.08
    åĿļæĮģ以
    -0.08
    POSITIVE LOGITS
    æĬķ票
    0.11
    positor
    0.09
    çļĦåĬ¨åĬĽ
    0.09
    essim
    0.08
     голоÑģ
    0.08
    ÑİÑĢ
    0.08
    对çħ§æ£ĢæŁ¥
    0.08
    æĬ¥åIJį
    0.08
    åĬłåĢį
    0.08
    urator
    0.08
    Activations Density 0.068%

    No Known Activations