© 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. 30-TRANSCODER-HP
    7. 95527
    Prev
    Next
    INDEX
    Explanations

    punctuation and connectors

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ernes
    -0.27
    enes
    -0.27
    ware
    -0.27
    å¿«æĬ¥
    -0.26
    å¾Ĵ
    -0.26
    Specifier
    -0.26
    annels
    -0.26
    á»ķi
    -0.26
    strate
    -0.25
    们
    -0.25
    POSITIVE LOGITS
     as
    0.30
     when
    0.30
    ;
    0.28
     whether
    0.28
    ,
    0.28
     during
    0.28
    .
    0.28
     -
    0.27
     again
    0.27
     with
    0.27
    Activations Density 1.869%

    No Known Activations