© 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. 109472
    Prev
    Next
    INDEX
    Explanations

    Extracts snippets of text

    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
     hint
    -0.32
     Hint
    -0.31
    hint
    -0.28
     preference
    -0.27
    Hint
    -0.26
    lider
    -0.25
     hints
    -0.25
    istan
    -0.25
    enever
    -0.25
     casual
    -0.24
    POSITIVE LOGITS
    ä¹Łä¸įæĥ³
    0.28
    iasco
    0.27
    éļ¾é¢ĺ
    0.25
    обоÑĢ
    0.25
    pon
    0.24
    GIT
    0.24
     Wanna
    0.23
     tras
    0.23
    丢失
    0.22
    ãģ¨æĢĿãģ£ãģŁ
    0.22
    Activations Density 0.124%

    No Known Activations