© 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. 11120
    Prev
    Next
    INDEX
    Explanations

    named after

    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
    atty
    -0.11
    otte
    -0.09
    è¿ĩåĪĨ
    -0.09
    æıŃç§ĺ
    -0.09
    çªŁ
    -0.08
    annes
    -0.08
    tone
    -0.08
    anka
    -0.08
    anni
    -0.08
    phans
    -0.08
    POSITIVE LOGITS
    een
    0.11
    å°ıåºĹ
    0.10
    æĮĩçĤ¹
    0.10
    /down
    0.10
    amp
    0.10
    noon
    0.09
    æ±Ģ
    0.09
    äºĭåIJİ
    0.09
    wards
    0.09
    åŃIJåŃĻ
    0.09
    Activations Density 0.023%

    No Known Activations