© 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. 58328
    Prev
    Next
    INDEX
    Explanations

    organizations and groups

    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
    lcd
    -0.09
    zilla
    -0.09
    å»ī
    -0.09
    çŀĭ
    -0.08
    setDescription
    -0.08
    æģĤ
    -0.08
    åĪ¬
    -0.08
    解æĶ¾æĢĿæĥ³
    -0.08
    å¾Ģ
    -0.08
    chie
    -0.08
    POSITIVE LOGITS
    åIJįåı«
    0.19
    åı«åģļ
    0.19
    åIJį为
    0.18
     called
    0.15
    called
    0.13
    称为
    0.13
    åı«
    0.13
    åij½åIJį为
    0.12
    Called
    0.10
    ä¹ĭä¸Ģ
    0.10
    Activations Density 0.380%

    No Known Activations