© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35581
    Prev
    Next
    INDEX
    Explanations

    to the

    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
    umm
    -0.10
    æĬĴ
    -0.10
    æĬĬè¿Ļ个
    -0.10
    åĪ°è¾¾
    -0.10
    RefPtr
    -0.09
    tica
    -0.09
    fragistics
    -0.09
    æĬĬ
    -0.09
    oldt
    -0.09
    ToAdd
    -0.09
    POSITIVE LOGITS
    gether
    0.13
    缮åīį为æŃ¢
    0.12
    诸å¦Ĥ
    0.11
    ä¸Ģå®ļçļĦ
    0.11
    æ¯ıä¸Ģ个人
    0.10
    çĤ¹äºĨ
    0.10
    ä¸Ģå®ļ
    0.09
    /from
    0.09
    åĪļæīį
    0.09
    åѦä¼ļäºĨ
    0.09
    Activations Density 0.746%

    No Known Activations