© 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. 58798
    Prev
    Next
    INDEX
    Explanations

    enjoying things

    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
    åĨ²åĬ¨
    -0.09
    æľīæķĪ
    -0.09
    овое
    -0.09
    好æĦŁ
    -0.08
    (QtCore
    -0.08
    ë°į
    -0.08
    æľīæľŁ
    -0.08
    æµļ
    -0.08
    æľŁæľ«
    -0.08
    åķī
    -0.08
    POSITIVE LOGITS
    ably
    0.13
    fully
    0.11
    ful
    0.09
    çľº
    0.09
    iest
    0.09
    год
    0.08
    çĶ¨èĩªå·±çļĦ
    0.08
    ited
    0.08
    æĪļ
    0.08
    able
    0.08
    Activations Density 0.011%

    No Known Activations