© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 88191
    Prev
    Next
    INDEX
    Explanations

    atis

    np_max-act · gemini-2.0-flash
    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
    Negative Logits
    hal
    -0.25
     mø
    -0.25
     general
    -0.25
     phÃŃa
    -0.24
    wright
    -0.24
    Do
    -0.24
    件
    -0.24
     hal
    -0.23
    psi
    -0.23
     remotely
    -0.23
    POSITIVE LOGITS
    åıĪèĥ½
    0.28
     sinks
    0.28
    æĶ«
    0.27
     CPP
    0.26
    cá
    0.26
     sinking
    0.26
    çļĦç«ŀäºī
    0.25
     snatch
    0.25
    chemas
    0.24
    Sink
    0.24
    Activations Density 5.434%

    No Known Activations