© 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. 21807
    Prev
    Next
    INDEX
    Explanations

    wired or Wired Magazine/data/network

    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
    åĽ¢éķ¿
    -0.10
    å¢Ł
    -0.09
    pek
    -0.09
    endants
    -0.09
     thủ
    -0.08
    pter
    -0.08
    overe
    -0.08
    ç¿Ĭ
    -0.08
    ÑĩеÑĤ
    -0.08
    .sap
    -0.08
    POSITIVE LOGITS
    ness
    0.11
    ician
    0.10
    æĬĢå¸Ī
    0.09
    æıĴ座
    0.09
    HAV
    0.09
    ï¸
    0.09
    稳妥
    0.08
    éĺ²çĽĹ
    0.08
     halt
    0.08
    екÑģ
    0.08
    Activations Density 0.006%

    No Known Activations