© 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. 141504
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    atron
    -0.33
     Loose
    -0.26
    æĮĩå®ļ
    -0.26
    éŁ³
    -0.26
    iÄħ
    -0.25
    åĿł
    -0.25
    ouston
    -0.24
    zÅij
    -0.24
    zhou
    -0.24
    uke
    -0.24
    POSITIVE LOGITS
     realism
    0.25
    æµį
    0.24
     proc
    0.24
    -ra
    0.24
     sorts
    0.24
    ewhat
    0.23
    ha
    0.23
    @interface
    0.23
    åĪ«
    0.23
    hea
    0.23
    Activations Density 15.686%

    No Known Activations

    This feature has no known activations.