© 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. 3648
    Prev
    Next
    INDEX
    Explanations

    leap, leaps, leapfrog, leap forward

    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
    éĺ¶å±Ĥ
    -0.10
    metic
    -0.09
    èĴº
    -0.09
    çĿ¬
    -0.08
    éħª
    -0.08
    ENTE
    -0.08
    ²
    -0.08
     Ballard
    -0.08
    ities
    -0.08
    uments
    -0.08
    POSITIVE LOGITS
    frog
    0.13
    edException
    0.10
    寻常
    0.09
    peare
    0.09
    ãĥ«ãĥī
    0.09
    acock
    0.09
    orde
    0.09
    ich
    0.09
    неÑĢ
    0.08
    ucht
    0.08
    Activations Density 0.005%

    No Known Activations