© 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. 17820
    Prev
    Next
    INDEX
    Explanations

    tre words

    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
    ames
    -0.09
    ä¸Ģ声
    -0.09
    script
    -0.08
    orm
    -0.08
    sWith
    -0.08
    sw
    -0.08
    robots
    -0.08
    éªĪ
    -0.08
    cv
    -0.08
    scan
    -0.08
    POSITIVE LOGITS
    egt
    0.12
     Ñģобой
    0.11
    stakes
    0.11
    ucht
    0.10
    housing
    0.10
    /Framework
    0.10
    EMBER
    0.10
    ег
    0.10
    viso
    0.09
    кол
    0.09
    Activations Density 0.020%

    No Known Activations