© 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. 58991
    Prev
    Next
    INDEX
    Explanations

    myjs.fr, myConnection, MyNetworkTV

    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
    ç»Ŀ对
    -0.09
    ãĥ¡
    -0.08
    ponsored
    -0.08
    .newBuilder
    -0.08
     Apprentice
    -0.08
    iva
    -0.08
     ØŃÙĪ
    -0.08
     Traverse
    -0.08
    ç»Ħè£ħ
    -0.08
    aison
    -0.08
    POSITIVE LOGITS
    马æĭī
    0.10
    åĶĽ
    0.09
    oom
    0.09
    åģļ大åģļ强
    0.09
     Malone
    0.08
    åĻ¤
    0.08
    hone
    0.08
    ìļ´ëıĻ
    0.08
    Ñĩик
    0.08
    çĽijçĿ£æ£ĢæŁ¥
    0.08
    Activations Density 0.076%

    No Known Activations