© 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. 33369
    Prev
    Next
    INDEX
    Explanations

    trend toward or trends in

    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
    ibr
    -0.09
    éŀ
    -0.09
     اÙĦØŃÙĬ
    -0.08
    çŃĶåį·
    -0.08
    å±ıéļľ
    -0.08
     ceil
    -0.08
    åī©
    -0.08
     led
    -0.08
    åIJĦå¼ıåIJĦæł·
    -0.08
    agus
    -0.08
    POSITIVE LOGITS
    è¶ĭåĬ¿
    0.35
     trends
    0.29
    çļĦè¶ĭåĬ¿
    0.28
     trend
    0.27
     Trends
    0.25
     Trend
    0.22
    åıijå±ķè¶ĭåĬ¿
    0.22
    æ½®æµģ
    0.19
    è¶ĭåIJij
    0.18
    è¶
    0.17
    Activations Density 0.102%

    No Known Activations