© 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. 36072
    Prev
    Next
    INDEX
    Explanations

    "ew" or "ek" followed by punctuation

    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
    ä¸Ģè¶Ł
    -0.10
    ä¸įä¹ħ
    -0.09
    è°İè¨Ģ
    -0.09
    çļĦç¼ĺæķħ
    -0.08
    ç¹ģåįİ
    -0.08
    çĤľ
    -0.08
    çļĦåŃĺåľ¨
    -0.08
    REN
    -0.08
    agnostic
    -0.08
    æijĨåľ¨
    -0.08
    POSITIVE LOGITS
    kes
    0.09
    ç´«å¤ĸ
    0.09
    à¹Īว
    0.08
    仪
    0.08
    enia
    0.08
    ively
    0.08
    utor
    0.08
    .Designer
    0.08
    abox
    0.08
    ht
    0.08
    Activations Density 0.004%

    No Known Activations