© 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. 4971
    Prev
    Next
    INDEX
    Explanations

    evaluation of

    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
    appen
    -0.10
    å®ŀå¹²
    -0.10
    ä¸įå®Į
    -0.09
     оÑĤлиÑĩа
    -0.09
     дела
    -0.09
     bó
    -0.09
    诱åıij
    -0.09
    ÇĴ
    -0.08
    诱导
    -0.08
     forts
    -0.08
    POSITIVE LOGITS
    è¯Ħä¼°
    0.13
    è¯Ħä»·
    0.12
    è¯Ħå®ļ
    0.11
    .evaluate
    0.11
    宽æĿ¾
    0.11
    çłĶåĪ¤
    0.10
    è©ķ
    0.10
     evaluate
    0.09
    å½¢åĬ¿
    0.09
    éªı
    0.09
    Activations Density 0.262%

    No Known Activations