© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37805
    Prev
    Next
    INDEX
    Explanations

    Dark Thirty, Close Encounters, Until Late

    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
    ä¹İ
    -0.11
    mania
    -0.10
    åŀĤ
    -0.09
    é£İå°ļ
    -0.09
     hob
    -0.09
    æį¢äºĨ
    -0.09
    _aliases
    -0.09
    croft
    -0.09
    IRC
    -0.08
    颢
    -0.08
    POSITIVE LOGITS
    èĴ²
    0.10
    -flight
    0.09
    大åݦ
    0.09
    æ¡ī
    0.08
    _gap
    0.08
    æĥł
    0.08
    ä»ħä¾Ľ
    0.08
     bre
    0.08
    ÙĪØ´
    0.08
    /component
    0.08
    Activations Density 0.001%

    No Known Activations