© 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. 39899
    Prev
    Next
    INDEX
    Explanations

    inches

    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
    çĹĩ
    -0.12
    åĴĮåľ°åĮº
    -0.11
    omic
    -0.09
    amat
    -0.09
    istical
    -0.09
    é£İå°ļ
    -0.09
    angler
    -0.08
    _checkpoint
    -0.08
    pez
    -0.08
     vicious
    -0.08
    POSITIVE LOGITS
    ernet
    0.12
    ãĥ³ãĤ°
    0.11
    ãĤħ
    0.11
    worm
    0.11
    ãĥ§ãĥ³
    0.10
    lers
    0.10
    累累
    0.09
    erman
    0.09
    apter
    0.09
    ord
    0.09
    Activations Density 0.013%

    No Known Activations