© 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. 14-TRANSCODER-HP
    7. 39434
    Prev
    Next
    INDEX
    Explanations

    uls

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    GMEM
    -0.27
    å®£ä¼łæ´»åĬ¨
    -0.26
    zem
    -0.26
    lay
    -0.26
    toFloat
    -0.25
    heid
    -0.25
     Sham
    -0.24
    _visibility
    -0.24
     Companion
    -0.24
    füg
    -0.24
    POSITIVE LOGITS
    arts
    0.28
     Depart
    0.26
    à¹Į
    0.25
    aper
    0.25
    aber
    0.24
     nucleus
    0.24
    ial
    0.24
    车éĺŁ
    0.24
     болÑĮ
    0.23
    ç£ķ
    0.23
    Activations Density 0.524%

    No Known Activations