© 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. 7910
    Prev
    Next
    INDEX
    Explanations

    methodology

    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
    ingly
    -0.10
    ENU
    -0.10
    ickle
    -0.09
    ikki
    -0.09
    鸳
    -0.09
    quired
    -0.08
    çıįè´µ
    -0.08
    weed
    -0.08
    çĭIJ
    -0.08
    apo
    -0.08
    POSITIVE LOGITS
    icc
    0.11
    æĢ§å¼º
    0.10
    /native
    0.08
    èĢĭ
    0.08
    ÑģÑĤан
    0.08
    æij©
    0.08
    ģ
    0.08
    رة
    0.08
    RITE
    0.08
    icho
    0.08
    Activations Density 0.025%

    No Known Activations