© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37811
    Prev
    Next
    INDEX
    Explanations

    software and technology solutions

    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
    pest
    -0.09
    æ¡£
    -0.09
    ìĸij
    -0.09
    èĤ¯
    -0.09
    dac
    -0.08
    éģĹæĨ¾
    -0.08
    é¢Ħå¤ĩ
    -0.08
    ãĥĥãĤ¯
    -0.08
    硬
    -0.08
    hea
    -0.08
    POSITIVE LOGITS
     suite
    0.11
    è¿Ļå¥Ĺ
    0.10
     core
    0.09
    åİŁçIJĨ
    0.09
    èĩªä¸»çłĶåıij
    0.09
    å¥Ĺ
    0.09
    åħ¨å¥Ĺ
    0.09
    .jupiter
    0.09
    suite
    0.09
    ç¬Ī
    0.09
    Activations Density 0.053%

    No Known Activations