© 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. 23355
    Prev
    Next
    INDEX
    Explanations

    proposal describes

    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
    lse
    -0.09
    bench
    -0.09
    ToOne
    -0.08
    _approx
    -0.08
    æ½Ń
    -0.08
    æľīæľºç»ĵåIJĪ
    -0.08
    åĪĪ
    -0.08
    çĨ¬
    -0.08
    auss
    -0.08
    询
    -0.08
    POSITIVE LOGITS
    æ¢Ń
    0.10
    éķ¶åµĮ
    0.09
    æľĢå°ij
    0.09
    eten
    0.09
    Mob
    0.09
    骰
    0.08
    éħĴ
    0.08
    iap
    0.08
    æľīèī²
    0.08
    åİĨ
    0.08
    Activations Density 0.396%

    No Known Activations