© 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. 50739
    Prev
    Next
    INDEX
    Explanations

    phrase

    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
    ãĤĥ
    -0.10
    ìį¨
    -0.09
    çĦ¶æĺ¯
    -0.09
    la
    -0.08
    /runtime
    -0.08
    olds
    -0.08
     ìŬ
    -0.08
    ajo
    -0.08
    roach
    -0.08
    åİ¿æĶ¿åºľ
    -0.08
    POSITIVE LOGITS
    æ»ĵ
    0.10
    Ħ
    0.10
    ä¹ĥèĩ³
    0.09
    enate
    0.09
    zcze
    0.09
    è¿Ļæł·çļĦè¯Ŀ
    0.09
    ıyla
    0.09
    wort
    0.09
    _MACRO
    0.09
    aus
    0.09
    Activations Density 0.009%

    No Known Activations