© 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. 32172
    Prev
    Next
    INDEX
    Explanations

    scare and scared

    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
    sonian
    -0.09
    ius
    -0.09
    ogn
    -0.09
    æĺ¥å¤ı
    -0.09
    bsp
    -0.09
    ertz
    -0.08
    ives
    -0.08
    ually
    -0.08
    cmds
    -0.08
    wers
    -0.08
    POSITIVE LOGITS
    crow
    0.12
    åĶ¬
    0.11
    aby
    0.11
    ä¸įå·²
    0.10
    åĩºå±Ģ
    0.09
    fully
    0.09
    åįİ侨
    0.09
    åħ±åIJĮä½ĵ
    0.09
    å¾Ĺå¾Ī
    0.09
    idge
    0.09
    Activations Density 0.009%

    No Known Activations