© 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. 20735
    Prev
    Next
    INDEX
    Explanations

    HP, capacity, output, single

    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
    潦
    -0.12
    äºĨä¸Ģåı¥
    -0.09
    æ¸İ
    -0.09
    鸡èĽĭ
    -0.09
    äºĶå®ĺ
    -0.09
    åı¯è¡Į
    -0.09
    éļıæīĭ
    -0.09
    çĥĤ
    -0.09
    éĴ¢ç»ĵæŀĦ
    -0.09
    itches
    -0.09
    POSITIVE LOGITS
    马åĬĽ
    0.10
    踵
    0.10
    åĬŁçİĩ
    0.10
    power
    0.10
    åĩ¡äºĭ
    0.10
    åıijåĬ¨æľº
    0.10
    1
    0.09
    éĽħ
    0.09
    5
    0.09
     моÑī
    0.09
    Activations Density 0.101%

    No Known Activations