© 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. 9418
    Prev
    Next
    INDEX
    Explanations

    thinking about

    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
    atty
    -0.11
    oll
    -0.09
    èĢĥèĻij
    -0.09
    åħ³æ³¨
    -0.09
    yt
    -0.09
    eus
    -0.09
    UTERS
    -0.09
     около
    -0.09
    enerating
    -0.09
    elson
    -0.09
    POSITIVE LOGITS
    ä»Ģä¹Īåij¢
    0.12
     how
    0.10
    ÙĪاÙĨ
    0.09
    tes
    0.09
    vido
    0.09
    /from
    0.09
    uts
    0.09
    ãģ®ãģ¯
    0.08
    ARI
    0.08
     the
    0.08
    Activations Density 0.069%

    No Known Activations