© 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. 22-TRANSCODER-HP
    7. 111640
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     wrestling
    -0.28
    `č↵
    -0.25
    &R
    -0.25
    orie
    -0.24
     'č↵
    -0.24
    edium
    -0.24
    æĦı大
    -0.23
     mast
    -0.23
    'H
    -0.23
     stad
    -0.23
    POSITIVE LOGITS
    è·µ
    0.28
    izando
    0.26
    ¨¡
    0.26
    ouser
    0.25
    ursal
    0.25
    för
    0.24
     cheeks
    0.24
    绿èī²éĢļéģĵ
    0.24
    æĸ
    0.24
    Ñģл
    0.24
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.