© 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. 8496
    Prev
    Next
    INDEX
    Explanations

    dioceses

    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
    é¦ĸ缸
    -0.10
    _stderr
    -0.09
    åĽ½åĬ¡
    -0.09
    çļĦç¡®æĺ¯
    -0.08
    åĽ½åº¦
    -0.08
    å¸Ĥå§Ķ常å§Ķ
    -0.08
    apore
    -0.08
    åĽ´
    -0.08
    emale
    -0.08
    enti
    -0.08
    POSITIVE LOGITS
    pecific
    0.12
    åĨħçļĦ
    0.11
    èĮĥåĽ´åĨħ
    0.10
    åĴĮ个人
    0.10
     haste
    0.10
    wide
    0.10
    -wide
    0.10
    èĩ´åĬĽ
    0.09
    -specific
    0.09
    plet
    0.09
    Activations Density 0.020%

    No Known Activations