© 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. 4230
    Prev
    Next
    INDEX
    Explanations

    leading to a

    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
    orque
    -0.09
    è¢Ń
    -0.09
    urous
    -0.08
    缬
    -0.08
    ifs
    -0.08
    寺
    -0.08
    æķijåij½
    -0.08
    è·ijäºĨ
    -0.08
    azel
    -0.08
    延éķ¿
    -0.08
    POSITIVE LOGITS
    åĨ³
    0.10
    çļĦå°±æĺ¯
    0.09
    渡
    0.08
    äºĨä¸Ģç³»åĪĹ
    0.08
    .Sync
    0.08
    ä¸įäºĨ
    0.08
    åħĭæľį
    0.08
    $rs
    0.08
    ÑįÑĤ
    0.08
    å¦Ĥä¸ĭ
    0.08
    Activations Density 0.006%

    No Known Activations