© 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. 107368
    Prev
    Next
    INDEX
    Explanations

    code and text mixed

    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
    uve
    -0.30
    åĽŃ
    -0.29
     domain
    -0.28
    ió
    -0.27
    ÑĮеÑĢ
    -0.26
    ÑĤÑĮ
    -0.26
    éī´åĪ«
    -0.25
    ume
    -0.25
    umi
    -0.24
    usat
    -0.24
    POSITIVE LOGITS
    matches
    0.30
    ected
    0.28
    oola
    0.27
    lain
    0.25
     chunks
    0.25
    .echo
    0.25
     diffé
    0.24
    _batches
    0.24
    achs
    0.24
    porn
    0.24
    Activations Density 33.018%

    No Known Activations