© 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. 56272
    Prev
    Next
    INDEX
    Explanations

    though followed by punctuation

    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
     zwar
    -0.10
    äºĨä¸įå°ij
    -0.10
    çĶļèĩ³è¿ĺ
    -0.10
    ť
    -0.10
    åŃIJéĩĮ
    -0.10
    èĢĮä¸Ķè¿ĺ
    -0.09
    à¹Į
    -0.09
    ä¸įèĤ¯
    -0.09
    izable
    -0.09
    éļıå¤Ħ
    -0.09
    POSITIVE LOGITS
    lems
    0.09
    richt
    0.09
    ski
    0.09
    prene
    0.09
    ylvania
    0.08
    ships
    0.08
    iras
    0.08
    resses
    0.08
    ynes
    0.08
    Wy
    0.08
    Activations Density 0.013%

    No Known Activations