© 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. 27198
    Prev
    Next
    INDEX
    Explanations

    decisive outcomes, routs, blowouts

    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
    è¾Ī
    -0.10
    .wr
    -0.09
    妥
    -0.08
    çļĦåij½è¿IJ
    -0.08
    zm
    -0.08
    寸
    -0.08
    æĦ¦
    -0.08
    é¡·
    -0.08
    æ¯ı人
    -0.08
    .truth
    -0.08
    POSITIVE LOGITS
    ingly
    0.10
    lah
    0.10
    ÃŃs
    0.09
    oth
    0.09
    pling
    0.09
    buster
    0.08
    åı²è¯Ĺ
    0.08
    äºĨä¸Ģåľº
    0.08
    å¥Ķ
    0.08
    gger
    0.08
    Activations Density 0.004%

    No Known Activations