© 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. 18511
    Prev
    Next
    INDEX
    Explanations

    programming languages

    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
    /providers
    -0.28
    segments
    -0.27
     segments
    -0.27
     divisions
    -0.27
     portions
    -0.25
     provinces
    -0.25
    CMS
    -0.25
    viders
    -0.25
     division
    -0.24
     provid
    -0.24
    POSITIVE LOGITS
    -paying
    0.31
    ç³Ł
    0.30
    wn
    0.29
    游
    0.26
     responsibly
    0.26
    toa
    0.25
    交
    0.25
    ock
    0.25
    缸äºĴ
    0.25
    wc
    0.24
    Activations Density 0.011%

    No Known Activations