© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38861
    Prev
    Next
    INDEX
    Explanations

    preliminary study or report

    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
    ify
    -0.10
    æĿ¼
    -0.09
    ified
    -0.09
    urv
    -0.09
    æıĦ
    -0.09
    çĺ¿
    -0.09
    agra
    -0.09
    æł¹æľ¬
    -0.09
    PropertyParams
    -0.09
    æĹĦ
    -0.09
    POSITIVE LOGITS
    åľ°å¸¦
    0.10
    quisite
    0.08
     stages
    0.08
    åĬŁè¯¾
    0.08
    reece
    0.08
    æ±°
    0.08
     noteworthy
    0.08
    æĮ¥
    0.08
    ģı
    0.08
    -stage
    0.08
    Activations Density 0.041%

    No Known Activations