© 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. 35297
    Prev
    Next
    INDEX
    Explanations

    degree, certainty, absolute

    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
     дело
    -0.11
    纷纷
    -0.10
    åIJĪéĢĤçļĦ
    -0.10
    çĶŁæ´»æĸ¹å¼ı
    -0.10
    -rating
    -0.09
    窥
    -0.09
    社ä¿Ŀ
    -0.09
    èĪĴéĢĤçļĦ
    -0.09
     Dom
    -0.09
     далеко
    -0.09
    POSITIVE LOGITS
    å®ģéĿĻ
    0.10
    ä¸ĵ注
    0.10
    éĢıæĺİ
    0.10
    紧迫
    0.10
    å¸ĮæľĽèĥ½
    0.09
    èĩªå¾ĭ
    0.09
    è¿Ļ份
    0.09
    é¢Ħè§ģ
    0.09
    便åĪ©
    0.09
    /widget
    0.09
    Activations Density 0.331%

    No Known Activations