© 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. 141523
    Prev
    Next
    INDEX
    Explanations

    Code and article snippets

    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
    pressions
    -0.28
    azel
    -0.27
    utive
    -0.25
    pard
    -0.25
    ãĤŃãĥ£ãĥ³
    -0.24
    ipheral
    -0.24
    Æ¡
    -0.24
     stay
    -0.23
    á»Ŀ
    -0.23
     patterns
    -0.23
    POSITIVE LOGITS
     princ
    0.26
    ulton
    0.26
    éŃį
    0.26
     welcomes
    0.25
     nan
    0.24
    ulfilled
    0.24
     instead
    0.24
    ulus
    0.24
     nightlife
    0.24
    placer
    0.24
    Activations Density 6.139%

    No Known Activations