© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52888
    Prev
    Next
    INDEX
    Explanations

    official source said

    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
    èĨĪ
    -0.09
    .servers
    -0.08
     orbs
    -0.08
     peri
    -0.08
    ãĤ¼
    -0.08
    ä¸Ī
    -0.08
     <![
    -0.07
     dul
    -0.07
    jak
    -0.07
    åĪ¹
    -0.07
    POSITIVE LOGITS
    åijĬè¯ī大家
    0.10
    éĢıéľ²
    0.10
    .datab
    0.10
    OSE
    0.09
    CADE
    0.09
    losed
    0.09
    è¿Ļé¦ĸæŃĮ
    0.09
    æĢħ
    0.09
    nop
    0.09
    ilden
    0.08
    Activations Density 0.092%

    No Known Activations