© 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. 88841
    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
    bred
    -0.30
     depend
    -0.27
    ród
    -0.26
    管çIJĨä¸Ńå¿ĥ
    -0.24
    anas
    -0.24
     demean
    -0.24
    æ¯Ķå¦Ĥ说
    -0.24
    orners
    -0.24
    è¯Ŀ说
    -0.24
    aned
    -0.24
    POSITIVE LOGITS
     sanction
    0.27
     sinking
    0.26
    ainless
    0.26
     region
    0.24
     Ce
    0.24
    åľ°åĮº
    0.23
    olla
    0.23
    mis
    0.23
    aghetti
    0.23
     маÑĤеÑĢиалÑĭ
    0.23
    Activations Density 10.151%

    No Known Activations

    This feature has no known activations.