© 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. 18415
    Prev
    Next
    INDEX
    Explanations

    Master of

    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
    ocz
    -0.09
    atre
    -0.09
    é»İæĺİ
    -0.08
    Ñĩик
    -0.08
    amous
    -0.08
    漪
    -0.08
    ÙĪØ©
    -0.08
    æľĿ
    -0.08
    едеÑĢ
    -0.08
    abin
    -0.08
    POSITIVE LOGITS
    pieces
    0.16
    fully
    0.14
    stroke
    0.13
    çĪ·
    0.13
    mind
    0.12
    éĶı
    0.12
    ëĭĺ
    0.12
    ship
    0.12
    们
    0.11
    ful
    0.11
    Activations Density 0.021%

    No Known Activations