© 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. 54473
    Prev
    Next
    INDEX
    Explanations

    duration or past time

    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
    elli
    -0.09
    们
    -0.09
    ä¸įäºĨ
    -0.09
    oid
    -0.09
    etheless
    -0.09
    eer
    -0.09
    ives
    -0.09
    rient
    -0.08
    ä¸Ģ種
    -0.08
    èµ·ä¾Ĩ
    -0.08
    POSITIVE LOGITS
     ago
    0.10
    工夫
    0.10
    د
    0.10
    媳
    0.09
    hound
    0.09
    å°¤åħ¶æĺ¯
    0.09
    çĶļèĩ³æĺ¯
    0.08
    ëį°
    0.08
    sburg
    0.08
    ago
    0.08
    Activations Density 0.012%

    No Known Activations