© 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. 58729
    Prev
    Next
    INDEX
    Explanations

    office of

    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
    媳
    -0.09
    viz
    -0.09
    hey
    -0.09
    è¾ĩ
    -0.08
    heart
    -0.08
    buff
    -0.08
    éľĵ
    -0.08
    åĪĹ
    -0.08
    pei
    -0.08
    wyn
    -0.08
    POSITIVE LOGITS
    å¿ħä¸įåı¯
    0.10
     Gone
    0.09
    UDO
    0.09
     khảo
    0.09
    aming
    0.08
    uate
    0.08
    uality
    0.08
    é¢ĺ主
    0.08
    avors
    0.08
    สรร
    0.08
    Activations Density 0.008%

    No Known Activations