© 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. 7558
    Prev
    Next
    INDEX
    Explanations

    battlefield contexts

    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
    asin
    -0.10
    梢
    -0.09
    ba
    -0.08
     Rebellion
    -0.08
     lift
    -0.08
    aim
    -0.08
     Demp
    -0.08
    件
    -0.08
    avel
    -0.08
    ounces
    -0.08
    POSITIVE LOGITS
    ä¸ĬçļĦ
    0.13
    -ready
    0.10
    累累
    0.10
    èĪĮ头
    0.10
    ä¸Ĭæľī
    0.10
    æĶ¾æīĭ
    0.09
    æĬŁ
    0.09
     дог
    0.09
    è¦ģä¹Ī
    0.09
     Ahead
    0.09
    Activations Density 0.028%

    No Known Activations