© 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. 27445
    Prev
    Next
    INDEX
    Explanations

    like or hate

    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
    anni
    -0.11
    äºĮåįģäºĶ
    -0.09
    综
    -0.08
    æŃ£åĢ¼
    -0.08
    åIJĮæĥħ
    -0.08
    ILLE
    -0.08
    å¸ĮæľĽ
    -0.08
    Callbacks
    -0.08
    æķĻèĤ²
    -0.08
    .minimum
    -0.08
    POSITIVE LOGITS
    è¾ĵäºĨ
    0.10
    åŃ¦æł¡çļĦ
    0.10
    ä¸į代表
    0.10
     пÑĢоведен
    0.09
    é£Ĵ
    0.09
     оÑģнован
    0.09
    çĹĤ
    0.09
    æĭ¿äºĨ
    0.09
    äºĽä»Ģä¹Ī
    0.09
    بÙĩ
    0.08
    Activations Density 0.021%

    No Known Activations