© 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. 23726
    Prev
    Next
    INDEX
    Explanations

    The Flash CW

    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
    俱
    -0.09
    èĦı
    -0.09
    -port
    -0.09
    .hot
    -0.09
    èİł
    -0.08
    大å±Ģ
    -0.08
    jeta
    -0.08
     port
    -0.08
    ä¼°
    -0.08
    çĬĬ
    -0.08
    POSITIVE LOGITS
    åİį
    0.09
    æīŀ
    0.09
    åı¦å¤ĸä¸Ģ个
    0.08
    	RE
    0.08
    çĪº
    0.08
    sr
    0.08
    çİĭèĢħ
    0.08
    alom
    0.08
     RESP
    0.08
    æİı
    0.08
    Activations Density 0.072%

    No Known Activations