© 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. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 108509
    Prev
    Next
    INDEX
    Explanations

    Scientific/analytical texts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    天天
    -0.31
    æ¯ıæĹ¥
    -0.28
    æ¯ı天
    -0.27
    常年
    -0.27
    æĽ´éĢĤåIJĪ
    -0.26
    åĮ®
    -0.26
    æĺŁçº§éħĴåºĹ
    -0.25
    å®ŀåĬ¡
    -0.25
    éĻIJ
    -0.25
    禧
    -0.24
    POSITIVE LOGITS
     unexpected
    0.30
     Scalars
    0.27
    [ch
    0.27
    éĤ£æ¬¡
    0.26
    ä»»ä½ķå½¢å¼ı
    0.26
    reation
    0.25
    /all
    0.25
    غÙĪ
    0.25
    åķIJ
    0.25
    åİĨæŶ
    0.24
    Activations Density 0.174%

    No Known Activations