© 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. 54367
    Prev
    Next
    INDEX
    Explanations

    pressing, pressure, and stamping

    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
    峪
    -0.10
     Scot
    -0.10
    ÑĢез
    -0.08
    æĬī
    -0.08
    å±ıéļľ
    -0.08
    ç§ijæĬĢåĽŃ
    -0.08
    çį»
    -0.08
    _compat
    -0.08
    ledge
    -0.08
    åįĩ
    -0.08
    POSITIVE LOGITS
    åİĭè¿«
    0.13
    å¡ij
    0.12
    åİĭ
    0.12
    å£ĵ
    0.11
    åİĭå®ŀ
    0.11
     Press
    0.11
    pressed
    0.11
     pressing
    0.11
    Press
    0.11
    åİĭ缩
    0.11
    Activations Density 0.119%

    No Known Activations