© 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. 55215
    Prev
    Next
    INDEX
    Explanations

    soils, profiles, cycled

    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
    大海
    -0.08
    ä¹ĭåŁİ
    -0.08
    åħ¨èº«
    -0.08
    个èĤ¡
    -0.08
    说ä¸įåĩº
    -0.08
    sip
    -0.08
    èį·
    -0.08
     bath
    -0.08
    ployment
    -0.07
     Rig
    -0.07
    POSITIVE LOGITS
    åľŁå£¤
    0.11
    èĢķ
    0.11
    avic
    0.11
    åĺĢ
    0.10
     поÑĩ
    0.10
    åŁĭ
    0.10
    æļ§
    0.10
    为空
    0.09
    .classes
    0.09
    æݼ
    0.09
    Activations Density 0.092%

    No Known Activations