© 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. 2403
    Prev
    Next
    INDEX
    Explanations

    hamradio.si, IV, Dong, .com, developers, FM, eWEEK, Platform

    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
    äºĭçī©
    -0.10
    åı®åĺ±
    -0.09
    ìķĮ
    -0.09
    æİİ
    -0.09
    åľĪ
    -0.09
    äºĶæĺŁ
    -0.09
    Ļ¨
    -0.09
    èĪħ
    -0.08
    æľºåħ³
    -0.08
    ä¸įè¿ľå¤Ħ
    -0.08
    POSITIVE LOGITS
    erin
    0.10
    ä¸įæŃ»
    0.09
     беÑĢем
    0.09
     DAM
    0.08
     Shak
    0.08
    èĻ¾
    0.08
    èĵ¼
    0.08
    ä¸ļä½Ļ
    0.08
    åľī
    0.08
    aths
    0.08
    Activations Density 0.010%

    No Known Activations