© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36972
    Prev
    Next
    INDEX
    Explanations

    before or prior to

    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
    ä¹ī
    -0.09
    trak
    -0.09
    acy
    -0.09
    влек
    -0.09
    èĹIJ
    -0.09
     diá»ħn
    -0.09
    ĺìĿ´
    -0.08
    Undo
    -0.08
    èĭ±éĽĦ
    -0.08
     Livingston
    -0.08
    POSITIVE LOGITS
    (before
    0.12
    çĿ¡åīį
    0.11
    before
    0.10
    åĵ¨
    0.10
     before
    0.10
    å°±è¯Ĭ
    0.10
     ÙĪبعد
    0.10
    æľŁåĨħ
    0.09
     BEFORE
    0.09
    磬
    0.09
    Activations Density 0.069%

    No Known Activations