© 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. 55586
    Prev
    Next
    INDEX
    Explanations

    apparatus and methods

    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
    isz
    -0.09
    utor
    -0.09
    ularity
    -0.08
    venue
    -0.08
    acious
    -0.08
    utors
    -0.08
    ucid
    -0.08
    åı¯æĥľ
    -0.08
    ummies
    -0.08
    اÙģÙĬ
    -0.08
    POSITIVE LOGITS
    lah
    0.12
    eer
    0.12
    ï¸ı
    0.10
    ively
    0.10
    zem
    0.10
    ting
    0.10
    ï¸
    0.09
    OfSize
    0.09
    æĥħ人èĬĤ
    0.09
    æĸ°æĪ¿
    0.09
    Activations Density 0.081%

    No Known Activations