© 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. 141227
    Prev
    Next
    INDEX
    Explanations

    states and directions

    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
    ugar
    -0.27
    upon
    -0.25
    两端
    -0.25
    åľ¨åĽ½å¤ĸ
    -0.24
    俱
    -0.24
    olla
    -0.23
    gh
    -0.23
    mods
    -0.23
    aturday
    -0.23
    quina
    -0.23
    POSITIVE LOGITS
    ®
    0.25
     Hera
    0.24
     ®
    0.24
    prene
    0.24
    ､
    0.23
    oce
    0.23
    ILI
    0.23
     Predictor
    0.23
    arians
    0.23
    .Invariant
    0.23
    Activations Density 18.313%

    No Known Activations