© 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. 62822
    Prev
    Next
    INDEX
    Explanations

    wetness or wet surfaces

    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
    eration
    -0.10
    ively
    -0.10
    ANCE
    -0.10
    èįīæ¡Ī
    -0.09
    orate
    -0.09
    身æĿIJ
    -0.09
    onto
    -0.09
    åŁĭ
    -0.09
    èijĨ
    -0.09
    èħ¹éĥ¨
    -0.09
    POSITIVE LOGITS
    æ¼ī
    0.12
    ting
    0.11
    åĩĩ
    0.11
    ening
    0.10
    /software
    0.10
    ted
    0.10
    apons
    0.10
    ters
    0.10
    tings
    0.10
    ensch
    0.10
    Activations Density 0.019%

    No Known Activations