© 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. 34044
    Prev
    Next
    INDEX
    Explanations

    piles of

    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
    åIJijä¸Ĭ
    -0.10
    æ´£
    -0.09
    æĹ©æĻļ
    -0.09
    ä¸ī级
    -0.09
    å®ļä½į
    -0.08
     bells
    -0.08
    ages
    -0.08
    帼
    -0.08
    çīµå¼ķ
    -0.08
    _HOOK
    -0.08
    POSITIVE LOGITS
    åłĨ积
    0.24
    åłĨ
    0.24
     pile
    0.19
     piles
    0.18
    åłĨæĶ¾
    0.18
    pile
    0.16
    ä¸ĢåłĨ
    0.15
     stack
    0.14
    æijŀ
    0.14
    积
    0.14
    Activations Density 0.065%

    No Known Activations