© 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. 34690
    Prev
    Next
    INDEX
    Explanations

    to the detriment or benefit

    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
     hảo
    -0.11
    æĢİä¹Īåı¯èĥ½
    -0.10
    æ£ļ
    -0.10
    èĭ
    -0.09
    çıį
    -0.09
     chances
    -0.08
     Interior
    -0.08
    è¹ļ
    -0.08
    象
    -0.08
    _Parameter
    -0.08
    POSITIVE LOGITS
     detriment
    0.13
     benefit
    0.11
    æŀģ大
    0.10
    æĪijå¾Ī
    0.10
    atisfaction
    0.09
    åѦçĶŁçļĦ
    0.09
    人å¿ĥ
    0.09
    éģĹæĨ¾
    0.09
    æĶ¶èİ·
    0.09
    åĽ°æĥij
    0.09
    Activations Density 0.016%

    No Known Activations