© 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. 55508
    Prev
    Next
    INDEX
    Explanations

    case of

    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
    æĻģ
    -0.10
    å¹ħ度
    -0.10
    éĹªçĶµ
    -0.09
    身边çļĦ
    -0.09
    colo
    -0.09
    ìĪĺ
    -0.09
    [++
    -0.08
    -shadow
    -0.08
    帽
    -0.08
    æĿ¥å½¢å®¹
    -0.08
    POSITIVE LOGITS
    URY
    0.11
    主管
    0.09
    æĪijæĺ¯
    0.09
    çľģåĨħ
    0.09
    禽
    0.09
    é¢Ĩ
    0.08
    istic
    0.08
    trag
    0.08
    å§Ĭ
    0.08
    èĪĤ
    0.08
    Activations Density 0.028%

    No Known Activations