© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6853
    Prev
    Next
    INDEX
    Explanations

    names ending in ard or ey

    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
    stab
    -0.09
    ANE
    -0.08
    æĪij们认为
    -0.08
     Saga
    -0.08
    nat
    -0.08
    æĪĸå¤ļ
    -0.08
    庵
    -0.08
    LOWER
    -0.08
    	pm
    -0.08
    èĢ¦
    -0.08
    POSITIVE LOGITS
    åħ¥ä½ı
    0.09
    annels
    0.09
    oose
    0.09
    ivation
    0.08
    darwin
    0.08
    omp
    0.08
    _FORE
    0.08
    åŁºå±Ĥ
    0.08
    å¿«ä¹IJ
    0.08
    åĩºè¡Į
    0.08
    Activations Density 0.003%

    No Known Activations