© 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. 20-TRANSCODER-HP
    7. 108231
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    èĻ½çĦ¶
    -0.32
    ä½Ĩ
    -0.32
    çĶļèĩ³
    -0.31
     but
    -0.30
     zwar
    -0.30
    ä½Ĩçİ°åľ¨
    -0.29
     но
    -0.28
     although
    -0.27
    ence
    -0.27
    èĻ½
    -0.27
    POSITIVE LOGITS
     nonetheless
    0.39
     nevertheless
    0.36
    并ä¸įæĺ¯
    0.34
    并éĿŀ
    0.32
    óż
    0.31
    ä»įçĦ¶æĺ¯
    0.28
    åį´æĺ¯
    0.27
    并ä¸į代表
    0.27
    并ä¸į
    0.27
     still
    0.26
    Activations Density 86.864%

    No Known Activations