© 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. 18-TRANSCODER-HP
    7. 89382
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lector
    -0.26
    ensitivity
    -0.24
    人æ°ijæľįåĬ¡
    -0.24
    åīĤ
    -0.24
    uke
    -0.23
    etc
    -0.23
    éĺ³
    -0.23
    è¨Ģè¡Į
    -0.23
     Pok
    -0.23
    RITE
    -0.23
    POSITIVE LOGITS
    OrDefault
    0.29
     dare
    0.26
    /single
    0.24
    Posting
    0.24
    adar
    0.24
     immediate
    0.24
    åİ¢
    0.24
     posting
    0.23
     Vern
    0.23
    ingle
    0.23
    Activations Density 67.173%

    No Known Activations

    This feature has no known activations.