© 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. 12-TRANSCODER-HP
    7. 23496
    Prev
    Next
    INDEX
    Explanations

    Mixed data sources

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uur
    -0.31
    .reddit
    -0.25
    ä¸ĢæĹ¥
    -0.25
     )"
    -0.25
    ATURE
    -0.25
     illustrates
    -0.24
    cies
    -0.24
    触åĬ¨
    -0.24
    è¡ĮåĪĹ
    -0.24
    çªģ
    -0.24
    POSITIVE LOGITS
    转移
    0.28
    seau
    0.28
    è¦ģåĬłå¼º
    0.27
     rubbed
    0.26
    ware
    0.25
    åºľ
    0.24
    opol
    0.24
    ford
    0.24
     mess
    0.24
    AndGet
    0.24
    Activations Density 0.044%

    No Known Activations