© 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. 19361
    Prev
    Next
    INDEX
    Explanations

    come arrival/occurrence

    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
    ï¸ı
    -0.11
    çľĭæľĽ
    -0.11
    ãģĦãģ¾ãģĻ
    -0.10
    ernes
    -0.10
    izz
    -0.10
    ãģĦãĤĭ
    -0.10
    çľĭ书
    -0.09
    ãģĦãģŁãĤī
    -0.09
    ãģĦãĤĭãģ¨
    -0.09
    enburg
    -0.09
    POSITIVE LOGITS
    culate
    0.11
    plete
    0.11
    place
    0.10
    好å¤Ħ
    0.09
     forth
    0.09
    bine
    0.09
     fruition
    0.09
     tắt
    0.09
    azon
    0.09
    ixer
    0.09
    Activations Density 0.068%

    No Known Activations