© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 107046
    Prev
    Next
    INDEX
    Explanations

    metric

    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
    Negative Logits
    ä»·ä½į
    -0.26
     surveyed
    -0.26
    å®ĺåı¸
    -0.25
     Florian
    -0.25
     Throughout
    -0.25
     anybody
    -0.23
    quake
    -0.23
    ctime
    -0.23
    wik
    -0.23
    ä»·æł¼ä¸Ĭ涨
    -0.23
    POSITIVE LOGITS
    еÑĢ
    0.26
    er
    0.25
    æī«
    0.25
    èĥ¨
    0.25
     fanc
    0.25
    念
    0.25
    ircular
    0.24
     mis
    0.24
    ä¾µåįł
    0.24
    iens
    0.23
    Activations Density 0.073%

    No Known Activations