© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10255
    Prev
    Next
    INDEX
    Explanations

    go

    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
    usz
    -0.10
    avel
    -0.08
    Ñģки
    -0.08
    lazy
    -0.08
    æĤ±
    -0.08
    edom
    -0.08
    æĿ¥ä¸´
    -0.08
    çĶŁæľº
    -0.08
    виÑĤ
    -0.08
    .getView
    -0.08
    POSITIVE LOGITS
    è¶Ł
    0.11
    <::
    0.09
    pawn
    0.08
    åĽ¢
    0.08
    .equalTo
    0.08
    EqualTo
    0.08
     nhu
    0.08
    ëĿ¼ìĿ´
    0.08
    æĬĬå®ĥ
    0.08
    thic
    0.08
    Activations Density 0.196%

    No Known Activations