© 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. 40362
    Prev
    Next
    INDEX
    Explanations

    Code and text mixture

    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
    ẩm
    -0.26
    Mt
    -0.25
    ward
    -0.25
    å¿Ĺ强
    -0.25
    -cloud
    -0.25
     lần
    -0.25
     unten
    -0.24
    èĻ¹æ¡¥
    -0.24
    urbed
    -0.24
    icted
    -0.24
    POSITIVE LOGITS
    alah
    0.26
    ÑĤеÑĢ
    0.25
    ascar
    0.25
     Gand
    0.25
    enso
    0.24
     hadn
    0.23
    æł¹æį®
    0.23
    æĻ®æ³ķ
    0.23
    aptor
    0.23
     Protest
    0.23
    Activations Density 0.049%

    No Known Activations