© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37271
    Prev
    Next
    INDEX
    Explanations

    defeat and victory

    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
    æĤ£
    -0.10
    ike
    -0.10
    archy
    -0.10
    ober
    -0.10
    _MISSING
    -0.09
     Fleming
    -0.09
    è´Łæĭħ
    -0.09
    çĭŃçªĦ
    -0.09
    kes
    -0.09
    ç»İ
    -0.09
    POSITIVE LOGITS
     defeat
    0.15
    æĮ«
    0.14
    è´¥
    0.14
    æķĹ
    0.14
    åĩ»è´¥
    0.13
    æīĵè´¥
    0.12
     defeated
    0.11
    失败
    0.11
    æ·ĺæ±°
    0.11
     defeats
    0.10
    Activations Density 0.062%

    No Known Activations