© 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. 62749
    Prev
    Next
    INDEX
    Explanations

    unless followed by condition

    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
     zwar
    -0.13
    -0.12
    ç»Īäºİ
    -0.11
    ä¹ĭæīĢ以
    -0.10
    èĻ½çĦ¶
    -0.09
    ä¸įä»ħåı¯ä»¥
    -0.09
    èĢĮä¸įæĺ¯
    -0.09
    zelf
    -0.09
    nt
    -0.09
    ase
    -0.09
    POSITIVE LOGITS
    æĺ¯åľ¨
    0.11
    /un
    0.11
    çī¹æ®ĬæĥħåĨµ
    0.11
    åı¦æľī
    0.10
    è¿Ļæł·æīįèĥ½
    0.10
    åħ¶
    0.10
    å¿ħè¦ģ
    0.09
    çĿĽ
    0.09
    æľīä»»ä½ķ
    0.09
    kus
    0.09
    Activations Density 0.024%

    No Known Activations