© 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. 10444
    Prev
    Next
    INDEX
    Explanations

    Montell, Montlake, Montserrat

    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
    jun
    -0.09
    leÅŁ
    -0.08
     moc
    -0.08
    éĨª
    -0.08
    ä¸Ī夫
    -0.08
    eph
    -0.08
    åĮĸçļĦ
    -0.07
     bó
    -0.07
    ONSE
    -0.07
    ogle
    -0.07
    POSITIVE LOGITS
    éªIJ
    0.10
    æŀģåĵģ
    0.09
    .descriptor
    0.09
    Ñīик
    0.09
    lli
    0.09
    æ¾į
    0.09
    orrent
    0.09
    _DR
    0.09
    çļĦ人çī©
    0.09
    å´Ľèµ·
    0.09
    Activations Density 0.045%

    No Known Activations