© 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. 23874
    Prev
    Next
    INDEX
    Explanations

    blend with

    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
    èĭ±éĽĦ
    -0.09
    åħ¸åŀĭ
    -0.09
    è·¨
    -0.09
    urable
    -0.08
    alon
    -0.08
    .microsoft
    -0.08
     вз
    -0.08
     Cummings
    -0.08
    ç»ĦåIJĪ
    -0.08
    èĢĹè´¹
    -0.08
    POSITIVE LOGITS
    è¡ĢèĦī
    0.12
    èŀį为ä¸Ģä½ĵ
    0.10
     surrounding
    0.10
    çľĭä¸įåĩº
    0.09
     existing
    0.09
    oka
    0.09
    engo
    0.09
     itself
    0.09
    èµ°ä¸ĭåİ»
    0.09
    èĩªèº«
    0.09
    Activations Density 0.044%

    No Known Activations