© 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. 35493
    Prev
    Next
    INDEX
    Explanations

    numbers followed by F.2d or different

    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
    eker
    -0.10
    ï¸ı
    -0.09
    ties
    -0.09
    棨
    -0.09
    IMUM
    -0.08
    ADER
    -0.08
     caval
    -0.08
    Ê°
    -0.08
    è¡«
    -0.08
    illed
    -0.08
    POSITIVE LOGITS
    çľĭä¸įæĩĤ
    0.11
    åĩĨæŶ
    0.09
    bsub
    0.09
    son
    0.08
    åħ¥ãĤĬ
    0.08
    als
    0.08
    swer
    0.08
     TForm
    0.08
    _NR
    0.08
    èŃ
    0.08
    Activations Density 0.173%

    No Known Activations