© 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. 40814
    Prev
    Next
    INDEX
    Explanations

    metabolism

    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
    alous
    -0.32
    egas
    -0.31
    uhl
    -0.25
    ÑģÑĤв
    -0.25
    ereo
    -0.25
    pleted
    -0.24
     flipped
    -0.24
     sterile
    -0.24
    ={}
    -0.24
    isé
    -0.24
    POSITIVE LOGITS
    åľ¨è¿ĩ渡
    0.26
    è´¨
    0.26
    overn
    0.26
    åŁºéĩijä¼ļ
    0.26
    æ³½
    0.26
    /libs
    0.26
     PRESS
    0.24
    赡
    0.24
    å¤ĩçĶ¨
    0.24
    .params
    0.23
    Activations Density 1.731%

    No Known Activations