© 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. 32070
    Prev
    Next
    INDEX
    Explanations

    many, one, few, different things

    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
    itest
    -0.10
    ä¸ĢåĪ»
    -0.10
    身价
    -0.09
    èħ¹
    -0.09
    ä¸Ģç»Ħ
    -0.09
    flate
    -0.09
    è®°å¿Ĩ
    -0.09
    .opens
    -0.09
    stasy
    -0.09
    ä¸įå°ıçļĦ
    -0.08
    POSITIVE LOGITS
     things
    0.27
     thing
    0.24
    things
    0.23
    äºĭæĥħ
    0.22
    ä¸ľè¥¿
    0.21
    çļĦäºĭæĥħ
    0.21
     Things
    0.20
    thing
    0.19
    çļĦäºĭ
    0.18
    ä¹ĭäºĭ
    0.18
    Activations Density 0.119%

    No Known Activations