© 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. 31176
    Prev
    Next
    INDEX
    Explanations

    numbers followed by month names

    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
    碰
    -0.08
    èĩº
    -0.08
     baÅŁÄ±na
    -0.08
    éºĭ
    -0.08
    æ¸
    -0.08
    ificados
    -0.08
    ponder
    -0.08
    rush
    -0.08
    æĪĸå¤ļ
    -0.08
    Å¡ÃŃ
    -0.08
    POSITIVE LOGITS
    eenth
    0.12
    teenth
    0.11
    th
    0.11
    een
    0.10
    è¿Ļ次
    0.10
    teen
    0.10
    æģĭæĥħ
    0.09
    antly
    0.09
    ä½IJ
    0.09
    å°ıé¾Ļ
    0.08
    Activations Density 0.378%

    No Known Activations