© 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. 22054
    Prev
    Next
    INDEX
    Explanations

    invention

    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
    wyn
    -0.10
    weit
    -0.09
    two
    -0.09
    FLAGS
    -0.09
    æĪĸå¤ļ
    -0.09
     impunity
    -0.09
     Ñģобой
    -0.09
    ByPrimaryKey
    -0.09
    æĤłæĤł
    -0.08
    è§ĦèĮĥåĮĸ
    -0.08
    POSITIVE LOGITS
    ä½ľçĶ¨
    0.10
    æĢ§çĸ¾çĹħ
    0.10
    ä¸Ģ线
    0.10
    åĬªåĬĽ
    0.09
    好转
    0.09
    pection
    0.09
    好å¥ĩ
    0.09
    verted
    0.09
    perial
    0.09
    amed
    0.09
    Activations Density 0.171%

    No Known Activations