© 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. 51906
    Prev
    Next
    INDEX
    Explanations

    stocking and populating

    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
    ulle
    -0.09
     filmy
    -0.08
     Lawn
    -0.08
    erro
    -0.08
    åĪĨæµģ
    -0.08
    itus
    -0.08
    ENC
    -0.08
    AWN
    -0.08
    sst
    -0.08
    åĮħåĽ´
    -0.08
    POSITIVE LOGITS
    ä¸ĢéĹ´
    0.10
    èĶĹ
    0.10
    èĨĽ
    0.10
    èµ·æĿ¥äºĨ
    0.09
    edImage
    0.09
    Ñĥк
    0.09
    iform
    0.09
    èIJģ
    0.09
    oppel
    0.08
    oret
    0.08
    Activations Density 0.009%

    No Known Activations