© 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. 19-TRANSCODER-HP
    7. 8606
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FIT
    -0.28
    _attachment
    -0.27
    (<?
    -0.27
    :<?
    -0.26
    ÑĨеп
    -0.26
    _CORE
    -0.25
    dsl
    -0.25
    cmc
    -0.25
    é¬ĺ
    -0.25
    dbo
    -0.24
    POSITIVE LOGITS
     leaking
    0.33
     leak
    0.31
     crushing
    0.28
     kiss
    0.28
     smell
    0.27
    ponible
    0.26
     drowning
    0.26
    unities
    0.25
     sensitive
    0.25
     dangerous
    0.25
    Activations Density 0.522%

    No Known Activations

    This feature has no known activations.