© 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. 54958
    Prev
    Next
    INDEX
    Explanations

    religious or spiritual terms

    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
     صØŃ
    -0.09
    çĬ
    -0.09
    åĪ¶æľį
    -0.08
     Äiju
    -0.08
    帮
    -0.08
     PERF
    -0.08
    æıIJæĭĶ
    -0.08
    lesi
    -0.08
    atism
    -0.08
    æ£į
    -0.08
    POSITIVE LOGITS
    åIJĪä¸Ģ
    0.10
     Kü
    0.08
     energ
    0.08
    æŃĮ声
    0.08
     discern
    0.08
    ÑĢÑıд
    0.08
    糯米
    0.08
    Mixin
    0.08
     egal
    0.08
    ä¹ĥæĺ¯
    0.07
    Activations Density 0.085%

    No Known Activations