© 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. 63664
    Prev
    Next
    INDEX
    Explanations

    can only verb

    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
     Golden
    -0.10
    uite
    -0.09
    Golden
    -0.09
    æĪ¾
    -0.09
     Squad
    -0.09
    æĶ¾è¿ĩ
    -0.08
    夫人
    -0.08
    ä¼ijåģĩ
    -0.08
    çĦī
    -0.08
    golden
    -0.08
    POSITIVE LOGITS
    åºĨ幸
    0.11
    以ä¸ĬçļĦ
    0.10
     HOR
    0.09
    åıªèĥ½è¯´
    0.09
    欣æħ°
    0.09
    ks
    0.09
     Ñģов
    0.09
    ç®Ĺæĺ¯
    0.09
    æĥ³ä¸įåĪ°
    0.08
    å½ĵä¹ĭ
    0.08
    Activations Density 0.011%

    No Known Activations