© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6798
    Prev
    Next
    INDEX
    Explanations

    reach out

    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
    éĢIJ
    -0.09
    åĶĩ
    -0.09
    æľī人
    -0.09
    赶
    -0.09
    fighters
    -0.08
    alle
    -0.08
     Ank
    -0.08
    _MI
    -0.08
    adle
    -0.08
    mares
    -0.08
    POSITIVE LOGITS
    anness
    0.11
    好åIJĹ
    0.09
    èĮ¬
    0.09
    让å®ĥ
    0.09
    äºĨä¸Ģä¸ĭ
    0.09
    åŃĺéĩı
    0.09
    为æĪij们
    0.08
    æŁ¥çľĭæĽ´å¤ļ
    0.08
    enberg
    0.08
     ÑģеÑĤи
    0.08
    Activations Density 0.011%

    No Known Activations