© 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. 58604
    Prev
    Next
    INDEX
    Explanations

    suffix -ian/-al/-ary/-er/-eur

    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
     usual
    -0.11
    ãģĨãģ¾ãģı
    -0.11
    å²ĸ
    -0.11
    çĿ£ä¿ĥ
    -0.10
     ÑĥÑĩаÑģÑĤ
    -0.10
    ÑģÑĥÑĤ
    -0.10
    º
    -0.10
    -quarters
    -0.10
    iable
    -0.10
    æĹłæķ°
    -0.10
    POSITIVE LOGITS
    itarian
    0.13
    窿
    0.13
    åĻľ
    0.12
    èĩ¾
    0.12
    iversal
    0.11
    ãģ¹ãģį
    0.11
    åĵĩ
    0.11
    iliary
    0.11
    ierrez
    0.10
    isateur
    0.10
    Activations Density 1.887%

    No Known Activations