© 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. 141720
    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
     collar
    -0.26
     cruc
    -0.26
     searchTerm
    -0.25
    виÑĤ
    -0.25
     NPC
    -0.25
    EEK
    -0.24
    xEE
    -0.24
    urat
    -0.23
     Vaults
    -0.23
    åĺ´
    -0.23
    POSITIVE LOGITS
    oran
    0.28
    è¡¡
    0.28
    fare
    0.25
    equality
    0.25
    棻
    0.25
    ...</
    0.24
    Wall
    0.24
    åīįåĪĹ
    0.24
    .tie
    0.24
    oding
    0.24
    Activations Density 4.707%

    No Known Activations

    This feature has no known activations.