© 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. 7979
    Prev
    Next
    INDEX
    Explanations

    scrapbook papers and layouts

    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
    è¯Ħæ¯Ķ
    -0.09
     pon
    -0.09
    æģ¶
    -0.09
    instruction
    -0.09
     wager
    -0.09
     Pon
    -0.09
    åħ´æĹº
    -0.08
    .Tool
    -0.08
    (instruction
    -0.08
    夫å¦ĩ
    -0.08
    POSITIVE LOGITS
    lei
    0.10
    åĭĭ
    0.09
    onders
    0.09
    æ¾İæ¹ĥ
    0.09
    ç¥ļ
    0.09
    æŀĦçŃij
    0.08
    ÑĤоÑĢ
    0.08
    翱
    0.08
    ä¸Ģå°ģä¿¡
    0.08
    éªĦåĤ²
    0.08
    Activations Density 0.015%

    No Known Activations