© 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. 19501
    Prev
    Next
    INDEX
    Explanations

    coff, coffee, coffers, coffin, Coffey

    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
    åľĥ
    -0.11
    è¹Ĭ
    -0.10
    åIJĮèĥŀ
    -0.09
     Jos
    -0.09
     ка
    -0.09
    ç«¿
    -0.09
    åĻľ
    -0.08
    é£İæĥħ
    -0.08
    å¿«æĬ¥
    -0.08
    åłĩ
    -0.08
    POSITIVE LOGITS
    lam
    0.10
    rell
    0.09
    mando
    0.09
    ridden
    0.08
     Compatibility
    0.08
    çľł
    0.08
    omo
    0.08
    èĥĻ
    0.08
    est
    0.08
    boxes
    0.08
    Activations Density 0.005%

    No Known Activations