© 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. 50381
    Prev
    Next
    INDEX
    Explanations

    fort or Fort followed by place names

    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
    åıī
    -0.10
    æīĴ
    -0.09
    uld
    -0.09
    éħļ
    -0.09
    æĿĪ
    -0.08
    åĪĨ管
    -0.08
    flo
    -0.08
     Nimbus
    -0.08
    ìľ
    -0.08
    inning
    -0.08
    POSITIVE LOGITS
    åħĭæĭī
    0.09
    çį»
    0.09
    singleton
    0.08
    åīĬå¼±
    0.08
     quot
    0.08
    .masks
    0.08
    asion
    0.08
    _PED
    0.08
    exo
    0.08
    typed
    0.08
    Activations Density 0.031%

    No Known Activations