© 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. 95540
    Prev
    Next
    INDEX
    Explanations

    counts and quantities

    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
     Roses
    -0.29
    æľīæľº
    -0.27
    ä»·
    -0.26
    prov
    -0.26
    åħĥç´ł
    -0.25
     bul
    -0.25
     elements
    -0.24
    Chelsea
    -0.24
    hes
    -0.23
    ä¸Ģèµ·æĿ¥
    -0.23
    POSITIVE LOGITS
    åĩłåįģ
    0.33
    åħ«åįģ
    0.29
     sixty
    0.29
    å»Ĭ
    0.28
    stead
    0.28
    åįģåħŃ
    0.28
     twenty
    0.28
     forty
    0.28
    ModelIndex
    0.27
    åįģåĩł
    0.27
    Activations Density 0.881%

    No Known Activations