© 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. 49363
    Prev
    Next
    INDEX
    Explanations

    warnings and alerts

    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
    ivated
    -0.10
    ç§ĥ
    -0.09
    ivation
    -0.09
    Poll
    -0.09
    éķľ
    -0.09
     Poll
    -0.08
    ivel
    -0.08
    _det
    -0.08
    ç¢Į
    -0.08
    antz
    -0.08
    POSITIVE LOGITS
    èѦåijĬ
    0.23
    èŃ¦ç¤º
    0.15
     warnings
    0.15
    é¢ĦèѦ
    0.15
     warning
    0.15
    warnings
    0.15
    .warn
    0.13
    warn
    0.13
     WARN
    0.13
     warn
    0.13
    Activations Density 0.073%

    No Known Activations