© 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. 31-TRANSCODER-HP
    7. 18387
    Prev
    Next
    INDEX
    Explanations

    Chemical compounds and code

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åħ¬åı¸ç«łç¨ĭ
    -0.28
    Persistence
    -0.27
    _Node
    -0.27
    åľ©
    -0.26
    ensis
    -0.24
    Node
    -0.24
    _node
    -0.24
    preter
    -0.24
    ôme
    -0.23
     AUX
    -0.23
    POSITIVE LOGITS
    asso
    0.32
    LAY
    0.29
    inst
    0.25
    æĴŃ
    0.25
     road
    0.25
     inst
    0.25
    /layouts
    0.24
    ãĥīãĥ¬ãĤ¹
    0.24
    asser
    0.24
     cushions
    0.24
    Activations Density 2.031%

    No Known Activations