© 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. 33206
    Prev
    Next
    INDEX
    Explanations

    about or at least

    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
     promin
    -0.09
    åĨįåĬłä¸Ĭ
    -0.09
    è¿ĺç®Ĺ
    -0.09
    зд
    -0.09
    幸好
    -0.08
    åĭī强
    -0.08
    æĺ¼å¤ľ
    -0.08
     roast
    -0.08
     Worcester
    -0.08
    äºĨä¸Ģ
    -0.08
    POSITIVE LOGITS
    ULE
    0.09
     Skywalker
    0.09
    _RS
    0.09
     COVER
    0.09
    .CreateInstance
    0.09
    çĩİ
    0.09
    æµģè¡Į
    0.09
    kers
    0.09
    è¦Ĩ
    0.09
    åħļåı²
    0.09
    Activations Density 0.015%

    No Known Activations