© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 62903
    Prev
    Next
    INDEX
    Explanations

    way, wonders, far

    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
    ULAR
    -0.10
    ulate
    -0.08
    æľīæķĪ
    -0.08
    fä
    -0.08
    StringEncoding
    -0.08
    _GENER
    -0.08
     Suitable
    -0.07
     nim
    -0.07
     griev
    -0.07
    panic
    -0.07
    POSITIVE LOGITS
    é¹ĥ
    0.11
    æīĵæī°
    0.09
    é£ı
    0.09
    0.09
    åĪĽæĬķ
    0.09
    ä½ĵçİ°äºĨ
    0.09
    ä»ŀ
    0.09
    мÑĭÑĪ
    0.09
    帮
    0.09
    å¼Ĥåľ°
    0.08
    Activations Density 0.004%

    No Known Activations