© 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. 58683
    Prev
    Next
    INDEX
    Explanations

    religious, faith, born-again, reform

    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
    缮çļĦåľ°
    -0.10
    ething
    -0.10
    redd
    -0.09
     Staff
    -0.09
    staff
    -0.08
    ÑĥлÑĮ
    -0.08
    alg
    -0.08
    amd
    -0.08
    олÑĮз
    -0.08
    çĦĻ
    -0.08
    POSITIVE LOGITS
     reform
    0.11
    æĶ¹éĿ©
    0.10
    èIJĮ
    0.09
    è¶Ĭæĺ¯
    0.09
    éĨĴ
    0.08
    æĶ¹éĿ©åĪĽæĸ°
    0.08
    éĿ¡
    0.08
     reforms
    0.08
    å±ħ室
    0.08
    .fig
    0.08
    Activations Density 0.065%

    No Known Activations