© 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. 55189
    Prev
    Next
    INDEX
    Explanations

    mantra, tantrum, tantra

    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
    ibri
    -0.11
    ëį°
    -0.10
    ä¸įèµ·
    -0.10
    纡
    -0.09
    OrElse
    -0.09
    stead
    -0.09
    crest
    -0.08
    irection
    -0.08
    èµ·æĿ¥
    -0.08
    åĿŁ
    -0.08
    POSITIVE LOGITS
    делÑĮ
    0.10
    à¹Į
    0.10
    æĢ§å¼º
    0.09
    æĥ³è±¡
    0.09
    opes
    0.09
    oes
    0.09
    ed
    0.09
    é»ĺé»ĺ
    0.08
    ä¸ĢåĪ»
    0.08
    å¾®ç¬ij
    0.08
    Activations Density 0.004%

    No Known Activations