© 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. 2712
    Prev
    Next
    INDEX
    Explanations

    story excerpt or highlights

    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
    è¿Ļåľº
    -0.11
    ç§ģ人
    -0.10
    è¯Ĺåı¥
    -0.09
    ç®´
    -0.09
    称ä¹ĭ
    -0.09
    é¹°
    -0.08
    éĥ½è¯´
    -0.08
    éĺľ
    -0.08
     Moran
    -0.08
    俳
    -0.08
    POSITIVE LOGITS
    æķħäºĭ
    0.15
     stories
    0.14
    Story
    0.13
    stories
    0.13
    çļĦæķħäºĭ
    0.13
    /story
    0.13
     story
    0.13
     Story
    0.12
    .story
    0.12
    -story
    0.12
    Activations Density 0.107%

    No Known Activations