© 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. 107196
    Prev
    Next
    INDEX
    Explanations

    Code, data, or formatting

    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
    èĺijèıĩ
    -0.31
    Tap
    -0.27
    .tap
    -0.27
    irm
    -0.26
    ingleton
    -0.25
    Mapped
    -0.25
    ServletRequest
    -0.25
    -b
    -0.24
    ectar
    -0.24
    UCK
    -0.24
    POSITIVE LOGITS
    çķ´
    0.29
     Gott
    0.26
    eldo
    0.26
    ]++;↵
    0.25
    ç½ij讯
    0.24
    \OptionsResolver
    0.24
    åijĹ
    0.24
    çİij
    0.24
     UserDetails
    0.24
    pst
    0.24
    Activations Density 0.107%

    No Known Activations