© 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. 21024
    Prev
    Next
    INDEX
    Explanations

    upside down, inverted, reversed

    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
    éĢĶ
    -0.12
     Gree
    -0.09
    gue
    -0.09
    unsch
    -0.09
     lan
    -0.09
    DTV
    -0.09
    åªĴ
    -0.08
    鼶è·Ŀ离
    -0.08
     Diy
    -0.08
    åĪĢ
    -0.08
    POSITIVE LOGITS
    é¢ł
    0.17
    éĢĨ转
    0.14
    éĢĨ
    0.14
     reversed
    0.13
    .reverse
    0.12
     upside
    0.12
     reversal
    0.11
     reverse
    0.11
    .Reverse
    0.11
    invert
    0.11
    Activations Density 0.104%

    No Known Activations