© 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. 57203
    Prev
    Next
    INDEX
    Explanations

    work 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
    é»ijæļĹ
    -0.10
    å¥ĭæĸĹ缮æłĩ
    -0.09
    dre
    -0.09
    anders
    -0.08
    èķĥ
    -0.08
     invent
    -0.08
    .generated
    -0.08
    ä¿Ń
    -0.08
    ieten
    -0.08
    ÑĢен
    -0.08
    POSITIVE LOGITS
    åı¯çĪ±çļĦ
    0.09
    ком
    0.08
    umen
    0.08
    ugin
    0.08
    á»ĩu
    0.08
    uder
    0.08
    ewn
    0.08
    ANGO
    0.08
    æĶ¿åįıå§Ķåijĺ
    0.08
    chein
    0.08
    Activations Density 0.026%

    No Known Activations