© 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. 35748
    Prev
    Next
    INDEX
    Explanations

    blurred lines

    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
    è°¤
    -0.09
    AffineTransform
    -0.09
    cour
    -0.09
    fest
    -0.09
    ÑĢоÑģÑĤ
    -0.08
    åĪ¶
    -0.08
    arov
    -0.08
     Tá»ķ
    -0.08
    mue
    -0.08
    å®ģåİ¿
    -0.08
    POSITIVE LOGITS
    模ç³Ĭ
    0.15
    çķĮéĻIJ
    0.15
    èŀį为ä¸Ģä½ĵ
    0.14
    ä¸įåĪĨ
    0.13
    æ·Ĩ
    0.13
    æ··æ·Ĩ
    0.13
     blurred
    0.12
    ä¸Ģä½ĵåĮĸ
    0.12
    ä¸Ģä½ĵ
    0.12
     blur
    0.11
    Activations Density 0.057%

    No Known Activations