© 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. 9293
    Prev
    Next
    INDEX
    Explanations

    elegant living, bad-ass, delicious steam, hard work, open source, spontaneous generation, long term, Home sewing, overt symbolism, clinical management

    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
    icum
    -0.11
     Affero
    -0.10
    éªİ
    -0.10
    管çIJĨä½ĵç³»
    -0.10
    å¹ķ
    -0.10
     маÑĤеÑĢиал
    -0.09
    steen
    -0.09
    Ñģов
    -0.09
    ä½ľäºĨ
    -0.09
    panse
    -0.09
    POSITIVE LOGITS
    è¿Ļ个è¯į
    0.09
     vs
    0.08
     being
    0.08
    è¿Ļ项
    0.08
     Eph
    0.08
     graft
    0.08
    tok
    0.08
    edic
    0.07
    ers
    0.07
    ek
    0.07
    Activations Density 0.434%

    No Known Activations