© 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. 108362
    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
    çĵ¶åŃIJ
    -0.25
    =settings
    -0.24
    addin
    -0.24
     (--
    -0.24
    åŃĺ
    -0.24
    ç´¯äºĨ
    -0.24
     Denn
    -0.24
     .....
    -0.24
     (?)
    -0.24
    .ExecuteNonQuery
    -0.24
    POSITIVE LOGITS
     поп
    0.31
    ika
    0.28
    _MORE
    0.27
    ikes
    0.26
     Links
    0.24
    管çIJĨåĪ¶åº¦
    0.24
    arkan
    0.24
    lies
    0.23
     Wyn
    0.23
    Replacing
    0.23
    Activations Density 2.784%

    No Known Activations

    This feature has no known activations.