© 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. 7816
    Prev
    Next
    INDEX
    Explanations

    opportunities

    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
    овÑĭй
    -0.10
    poi
    -0.09
    strom
    -0.09
    rock
    -0.09
    ld
    -0.08
    çħ
    -0.08
    ä»Ģä¹Īåı«
    -0.08
    åĵĪå°Ķ
    -0.08
    ä¾Ľåħ»
    -0.08
    ır
    -0.08
    POSITIVE LOGITS
     ways
    0.10
    acco
    0.10
     ÑĤек
    0.09
    åĴĮåıijå±ķ
    0.09
    渺
    0.09
    ä¸įå¤ļ
    0.09
    åħ±åIJĮä½ĵ
    0.08
    ноÑģÑĤей
    0.08
    åľ¨å®¶
    0.08
    /disable
    0.08
    Activations Density 0.020%

    No Known Activations