© 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. 36901
    Prev
    Next
    INDEX
    Explanations

    preferred choice

    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
    åıĹçĽĬ
    -0.10
    ä¸ĵèģĮ
    -0.10
    å²Ń
    -0.10
    åģ¥åħ¨
    -0.09
    çĥĽ
    -0.09
     Benef
    -0.09
    æĪĹ
    -0.09
    鸬
    -0.09
    stre
    -0.08
    æ³ķè§Ħ
    -0.08
    POSITIVE LOGITS
    é¦ĸéĢī
    0.16
    éĿĴçĿIJ
    0.13
    ä¿¡èµĸ
    0.13
     preferred
    0.12
     Preferred
    0.11
     choice
    0.11
    -choice
    0.10
    choice
    0.10
    çļĦéĢīæĭ©
    0.10
    åħĨ
    0.10
    Activations Density 0.077%

    No Known Activations