© 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. 22142
    Prev
    Next
    INDEX
    Explanations

    hardware and components

    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
    otec
    -0.11
    igraphy
    -0.09
    èĵ¼
    -0.09
    etwork
    -0.09
    å´Ľ
    -0.09
    ennen
    -0.09
    ively
    -0.09
    ulta
    -0.09
     Duffy
    -0.09
    ipi
    -0.08
    POSITIVE LOGITS
    bars
    0.09
     Dominion
    0.08
    家ä½ı
    0.08
    çĺķ
    0.08
    åģĮ
    0.07
    lov
    0.07
     Greatest
    0.07
    èľī
    0.07
    æļĤ
    0.07
    oenix
    0.07
    Activations Density 0.003%

    No Known Activations