© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40027
    Prev
    Next
    INDEX
    Explanations

    never expected

    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
    Ñģим
    -0.09
    æīİ
    -0.09
    çīĪæĿĥå½Ĵ
    -0.08
    ç²¾è£ħ
    -0.08
    ä¸Ģéĺµ
    -0.08
    ä¸Ģä½į
    -0.08
    碴
    -0.08
    åĽ´
    -0.08
    èĪ«
    -0.08
    饬
    -0.08
    POSITIVE LOGITS
    没æĥ³åĪ°
    0.21
    没æľīæĥ³åĪ°
    0.20
    æĥ³ä¸įåĪ°
    0.18
    é¢ĦæĸĻ
    0.14
    æĦıæĥ³ä¸įåĪ°
    0.14
     expectation
    0.13
    expect
    0.12
    Expect
    0.12
     expectations
    0.12
    åģļ梦
    0.12
    Activations Density 0.035%

    No Known Activations