© 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. 39017
    Prev
    Next
    INDEX
    Explanations

    gay, fruity, dramatic, fragile characters

    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
    夫å¦ĩ
    -0.10
    èĤĺ
    -0.09
    æľ´
    -0.09
    еÑħ
    -0.08
    æĭĨ
    -0.08
     Matthias
    -0.08
    å·¥ä½ľå®¤
    -0.08
    ESH
    -0.08
    ä¸Ńå¤ĸ
    -0.08
    _cast
    -0.08
    POSITIVE LOGITS
    æŁĶ
    0.10
    orno
    0.10
    åį§
    0.10
    å²±
    0.09
    å£
    0.09
    ä»ĸå¦Ī
    0.09
    å¯Įè£ķ
    0.09
    éĺī
    0.09
    é«ĺå°ļ
    0.09
    apis
    0.09
    Activations Density 0.107%

    No Known Activations