© 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. 54390
    Prev
    Next
    INDEX
    Explanations

    in an ideal position

    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
    5
    -0.09
    0
    -0.09
    ulet
    -0.09
    3
    -0.08
    9
    -0.08
    1
    -0.08
    Pet
    -0.08
    æĦ£
    -0.08
     Pet
    -0.08
    ä¸ŀ
    -0.08
    POSITIVE LOGITS
    æĽ´å®¹æĺĵ
    0.10
    [position
    0.10
     poised
    0.09
    æĽ´éĢĤåIJĪ
    0.09
    æľīæľĽ
    0.09
     mieux
    0.09
    _positions
    0.09
    (position
    0.09
    æľ
    0.09
    pared
    0.09
    Activations Density 0.029%

    No Known Activations