© 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. 56214
    Prev
    Next
    INDEX
    Explanations

    spent my lifespent more timespent the lastspent enough timespent the betterspent about an hourspent nearly allspent her early childhoodspent her teenage yearsspent most of my life

    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
    ../../../
    -0.12
    èĩ³äºİ
    -0.11
    INDER
    -0.09
    _OVERFLOW
    -0.09
    usk
    -0.09
    croll
    -0.09
     sized
    -0.09
    ìļ©
    -0.09
    çĴľ
    -0.08
     ÑģвÑıзи
    -0.08
    POSITIVE LOGITS
    å¾Īéķ¿æŶéĹ´
    0.13
    zes
    0.10
    iture
    0.10
    Ìģ
    0.10
    ä¸įæĸŃçļĦ
    0.09
    eacher
    0.09
     considerable
    0.09
    ifix
    0.09
    ablish
    0.08
    allet
    0.08
    Activations Density 0.015%

    No Known Activations