© 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. 14-TRANSCODER-HP
    7. 40593
    Prev
    Next
    INDEX
    Explanations

    patch

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.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
    ellig
    -0.27
     like
    -0.26
    ake
    -0.26
    ÅŁÄ±
    -0.26
    strike
    -0.25
     spot
    -0.25
      ↵
    -0.25
    ç¬ijäºĨ
    -0.24
    king
    -0.24
    POSITIVE LOGITS
    gie
    0.35
     harder
    0.29
    blers
    0.28
    Hard
    0.25
     hardest
    0.25
    ãĤīãģı
    0.25
    /apt
    0.24
     Fact
    0.24
     Ren
    0.24
     Modi
    0.24
    Activations Density 0.309%

    No Known Activations