© 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. 5667
    Prev
    Next
    INDEX
    Explanations

    feudalism, lords, fief, rivalry

    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
    ãģĵãģ¨
    -0.10
    acas
    -0.09
    .GPIO
    -0.09
    alom
    -0.09
    صاÙĦ
    -0.09
    èĵ¦
    -0.09
    éĻIJ度
    -0.09
    ÙĤÙĨ
    -0.08
     Mons
    -0.08
     Pick
    -0.08
    POSITIVE LOGITS
    ç³ľ
    0.09
    onne
    0.09
    åĴ§
    0.09
    ãģ¾ãģ¾
    0.08
    baz
    0.08
    hammer
    0.08
    æĹ¶ä»£
    0.08
    æ´¾
    0.08
    çĤ³
    0.08
     Crest
    0.08
    Activations Density 0.004%

    No Known Activations