© 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. 89698
    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
    .TestCase
    -0.28
    è¶ĬæĿ¥è¶Ĭå¤ļ
    -0.26
    -tw
    -0.26
     inflation
    -0.25
     Quantity
    -0.25
    ä¸Ģ度
    -0.24
     flap
    -0.24
    ç»ıåİĨè¿ĩ
    -0.24
     worn
    -0.24
    Quantity
    -0.24
    POSITIVE LOGITS
     zwar
    0.28
    peat
    0.28
    mdb
    0.27
    åĬ©
    0.26
    positor
    0.25
    lace
    0.25
    pci
    0.25
    iger
    0.25
     aprove
    0.24
    ÄĻb
    0.24
    Activations Density 1.956%

    No Known Activations

    This feature has no known activations.