© 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. 55152
    Prev
    Next
    INDEX
    Explanations

    contrast, in contrast, contrasts

    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
    çģµæ´»æĢ§
    -0.15
    åľľ
    -0.11
    çģµæ´»
    -0.10
     flexibility
    -0.10
     flexible
    -0.10
     Flex
    -0.09
    æŀŀ
    -0.09
    .flex
    -0.09
    è¿ĺæľīä»Ģä¹Ī
    -0.09
    flex
    -0.09
    POSITIVE LOGITS
     contrast
    0.23
    对æ¯Ķ
    0.22
    contrast
    0.19
     Contrast
    0.19
     contrasts
    0.16
     stark
    0.14
     CONTR
    0.14
    缸æ¯Ķä¹ĭä¸ĭ
    0.14
     contrasting
    0.13
    Contr
    0.13
    Activations Density 0.022%

    No Known Activations