© 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. 32302
    Prev
    Next
    INDEX
    Explanations

    affiliate compensation or commission

    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
    åħįè´¹
    -0.10
    åħįè²»
    -0.09
    涤
    -0.09
    妫
    -0.09
    ä¹°åĪ°
    -0.09
    åħ¬çĽĬ
    -0.09
    详
    -0.09
    å½¢æĪIJçļĦ
    -0.08
    ä¹°çļĦ
    -0.08
    геÑĢ
    -0.08
    POSITIVE LOGITS
    æĸĻ
    0.11
    æĦıåij³
    0.08
    ophilia
    0.08
    пÑĢоÑģ
    0.08
    æĹ¢è¦ģ
    0.08
    imag
    0.08
    ORY
    0.08
    èݪ
    0.08
     ingredients
    0.08
    æĪIJ交
    0.08
    Activations Density 0.018%

    No Known Activations