© 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. 111476
    Prev
    Next
    INDEX
    Explanations

    Document snippets

    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
     [--
    -0.28
    bish
    -0.26
    ære
    -0.26
    大ãģįãģĦ
    -0.24
    åħ³éĶ®æĬĢæľ¯
    -0.24
    ufs
    -0.24
    æıIJ款
    -0.24
    ",-
    -0.24
    wel
    -0.23
    æĢ§çļĦ
    -0.23
    POSITIVE LOGITS
    isActive
    0.30
     Paste
    0.28
    Priv
    0.27
     Priv
    0.27
    .io
    0.26
     Mai
    0.26
    -
    0.24
    辩
    0.24
    M
    0.24
     Ost
    0.24
    Activations Density 0.016%

    No Known Activations