© 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. 9550
    Prev
    Next
    INDEX
    Explanations

    always, forever, never

    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
    8
    -0.09
    ç»ıè´¹
    -0.08
    رÙĬد
    -0.08
     dit
    -0.08
    3
    -0.08
    stor
    -0.08
    0
    -0.08
    6
    -0.08
     Nes
    -0.08
    5
    -0.08
    POSITIVE LOGITS
    æ°¸è¿ľ
    0.18
    æ°¸éģł
    0.15
     forever
    0.15
     mãi
    0.12
     Forever
    0.12
    çĴĭ
    0.11
    æ°¸è¿ľä¸įä¼ļ
    0.10
    äºĮèĢħ
    0.10
    WSC
    0.10
    fore
    0.10
    Activations Density 0.029%

    No Known Activations