© 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. 49047
    Prev
    Next
    INDEX
    Explanations

    tumbled, umble, tumor

    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
    夷
    -0.09
     Epstein
    -0.09
    inne
    -0.09
    碣
    -0.08
    禮
    -0.08
    odom
    -0.08
    asm
    -0.08
     ÑĢÑĥк
    -0.08
    禳
    -0.08
    lessly
    -0.08
    POSITIVE LOGITS
    æ»ļæ»ļ
    0.10
    ç¢Į
    0.09
    rolling
    0.09
    澪
    0.09
    æ»ļåĬ¨
    0.09
    èĩªæŁ¥
    0.09
    æĶ¾åѦ
    0.09
    ÙĪØŃ
    0.09
    åıijçĶŁåľ¨
    0.08
    roll
    0.08
    Activations Density 0.008%

    No Known Activations