© 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. 64162
    Prev
    Next
    INDEX
    Explanations

    Past-tense verbs

    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
    .receive
    -0.28
    çķĢ
    -0.28
    eligible
    -0.25
    lip
    -0.24
    ungan
    -0.24
    given
    -0.24
     SelectList
    -0.24
     gemacht
    -0.24
    åĪĩéĻ¤
    -0.23
    ãģķãĤĮãģ¾ãģĹãģŁ
    -0.23
    POSITIVE LOGITS
     Died
    0.26
    åĴĮå¹³
    0.25
     dropped
    0.24
    audi
    0.24
    www
    0.24
    LAY
    0.23
     departed
    0.23
     depart
    0.23
    .printStackTrace
    0.23
    æķ°é¢Ŀ
    0.23
    Activations Density 15.387%

    No Known Activations