© 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. 27656
    Prev
    Next
    INDEX
    Explanations

    juju, magic, craft, oodoo

    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
    osal
    -0.09
    strup
    -0.09
    æĶĺ
    -0.09
     вÑģÑĤÑĢ
    -0.08
    gos
    -0.08
    atego
    -0.08
    inho
    -0.08
    ÖĤ
    -0.08
    çĹŀ
    -0.08
    wert
    -0.08
    POSITIVE LOGITS
    pras
    0.11
    çıij
    0.10
    èĭ´
    0.10
    ç©·
    0.09
    çķĮçļĦ
    0.09
    åŁ´
    0.09
    æĥħ人èĬĤ
    0.09
    ä¸īäºļ
    0.09
    ilor
    0.09
    ingleton
    0.09
    Activations Density 0.001%

    No Known Activations