© 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. 22-TRANSCODER-HP
    7. 97856
    Prev
    Next
    INDEX
    Explanations

    Math and Armenian

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     gust
    -0.29
    .jar
    -0.27
    .gridx
    -0.27
    uckle
    -0.26
    çĥŁ
    -0.26
     Cond
    -0.24
    aday
    -0.24
    ator
    -0.24
     Kaplan
    -0.24
     laut
    -0.24
    POSITIVE LOGITS
    lessly
    0.29
    erts
    0.28
    伶
    0.27
     Armenian
    0.26
     Anonymous
    0.26
    arry
    0.25
    Triangles
    0.25
    .stopPropagation
    0.24
    æĹ¶æľº
    0.24
    dex
    0.24
    Activations Density 0.063%

    No Known Activations