© 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. 35648
    Prev
    Next
    INDEX
    Explanations

    goal of

    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
    edImage
    -0.10
    oine
    -0.09
    loo
    -0.09
    Ñĥк
    -0.09
    aly
    -0.09
    enza
    -0.09
    ÑĮÑİ
    -0.09
    ature
    -0.09
    ÙģاÙĤ
    -0.09
    plorer
    -0.08
    POSITIVE LOGITS
    keepers
    0.12
    KEEP
    0.10
    inski
    0.10
    keeper
    0.10
    lessly
    0.09
     dÄ±ÅŁÄ±
    0.09
    itud
    0.09
    ä¹ĭåĨħ
    0.09
    odyn
    0.09
    /target
    0.08
    Activations Density 0.045%

    No Known Activations