© 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. 39628
    Prev
    Next
    INDEX
    Explanations

    back to the

    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
    æĬµ
    -0.09
    -away
    -0.09
    trag
    -0.09
     ë¡
    -0.09
    awl
    -0.09
    æµ´
    -0.09
    æīĺ
    -0.08
    Õ»
    -0.08
    ç»Īæŀģ
    -0.08
     radio
    -0.08
    POSITIVE LOGITS
    åĪļæīį
    0.10
    _COMPLEX
    0.09
    åİŁæĿ¥çļĦ
    0.09
     Democrat
    0.09
    åİŁæĿ¥
    0.09
    常æĢģ
    0.08
    éĥ¨èIJ½
    0.08
    æĦ¦
    0.08
    åİŁä¾Ĩ
    0.08
    oven
    0.08
    Activations Density 0.035%

    No Known Activations