© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7931
    Prev
    Next
    INDEX
    Explanations

    probe, murky, immersed, for, of, with, spill, facility, one, rubble

    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
     Fare
    -0.10
    eration
    -0.10
    >Type
    -0.09
     kısı
    -0.09
    çįĹ
    -0.08
     Released
    -0.08
     Sexy
    -0.08
    TargetException
    -0.08
    æłĩå¿ĹæĢ§
    -0.08
    fare
    -0.08
    POSITIVE LOGITS
    å±Ī
    0.10
    åľ¨æĦı
    0.09
    趣
    0.09
    èĪĨ论
    0.09
    å°±æĥ³
    0.09
    à¸IJ
    0.08
    人家
    0.08
     pr
    0.08
    åŃĸ
    0.08
    室å¤ĸ
    0.08
    Activations Density 0.576%

    No Known Activations