© 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. 34121
    Prev
    Next
    INDEX
    Explanations

    author:

    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
    å¹³åĿĩæ°´å¹³
    -0.08
     khúc
    -0.08
    ned
    -0.08
    VER
    -0.08
    .methods
    -0.08
    平衡
    -0.08
    é²ľèī³
    -0.07
    樱èĬ±
    -0.07
    laden
    -0.07
    avers
    -0.07
    POSITIVE LOGITS
    ware
    0.10
    Ñħи
    0.09
    èĶº
    0.09
    ovic
    0.09
    olis
    0.08
    庶
    0.08
    åIJĦ大
    0.08
    ä¼ī
    0.08
     unn
    0.08
     Falling
    0.08
    Activations Density 0.005%

    No Known Activations