© 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. 36495
    Prev
    Next
    INDEX
    Explanations

    command line

    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
    ening
    -0.11
    ï¸
    -0.11
    atto
    -0.10
    unny
    -0.10
    DRV
    -0.09
    ady
    -0.09
    uish
    -0.09
    olut
    -0.09
    getNode
    -0.09
    ä¿
    -0.08
    POSITIVE LOGITS
    wealth
    0.11
    大åħ¨
    0.11
    aris
    0.10
    -line
    0.10
    ä¸ĭè¾¾
    0.09
    aries
    0.09
    æݧèĤ¡éĽĨåĽ¢
    0.08
    ìĸ´
    0.08
    .CommandText
    0.08
    碼
    0.08
    Activations Density 0.083%

    No Known Activations