© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51294
    Prev
    Next
    INDEX
    Explanations

    melting wax or candles

    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
     mates
    -0.09
    箧
    -0.09
     Gilles
    -0.09
    onas
    -0.09
    æµĵæµĵ
    -0.08
     доп
    -0.08
    artz
    -0.08
    QUIT
    -0.08
    zi
    -0.08
    ç¿»å¼Ģ
    -0.08
    POSITIVE LOGITS
     Wax
    0.12
     wax
    0.12
    èľ¡
    0.12
    èŀįåĮĸ
    0.11
     melt
    0.09
     candles
    0.09
    åĽºåĮĸ
    0.09
    纬
    0.09
    ä¸ĢåĪĢ
    0.09
    åĬłçĽĸ
    0.09
    Activations Density 0.110%

    No Known Activations