© 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. 129914
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    utors
    -0.26
     Transcript
    -0.25
    éļ¾å¾Ĺ
    -0.25
    å°±æĥ³
    -0.25
    åŃ¦ä¹ł
    -0.24
    .Stat
    -0.24
     Redistributions
    -0.24
    mitters
    -0.24
    unce
    -0.23
    aines
    -0.23
    POSITIVE LOGITS
    æīį
    0.25
     carne
    0.24
    means
    0.23
     fate
    0.23
     ì²ĺ
    0.23
    å¤ĩçĶ¨
    0.23
    Backing
    0.23
    factory
    0.22
    _factory
    0.22
     qualifies
    0.22
    Activations Density 7.886%

    No Known Activations

    This feature has no known activations.