© 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. 20-TRANSCODER-HP
    7. 107416
    Prev
    Next
    INDEX
    Explanations

    exclusively/solely

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hta
    -0.26
    iverse
    -0.25
    inox
    -0.24
    ය
    -0.24
    å¼Ģå°ģ
    -0.24
    etti
    -0.24
    imps
    -0.24
    enh
    -0.23
    ä¸ŃéĹ´
    -0.23
    ä¸ĭåįĬ
    -0.23
    POSITIVE LOGITS
    _slice
    0.29
    odial
    0.27
     branded
    0.25
    åĩºå±Ģ
    0.25
    æµģè¡ĮçļĦ
    0.24
    Ĥ¬
    0.24
    -grade
    0.24
    assistant
    0.24
     sal
    0.24
    訾
    0.24
    Activations Density 18.060%

    No Known Activations