© 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. 88293
    Prev
    Next
    INDEX
    Explanations

    HTML code

    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
     punches
    -0.31
    Success
    -0.27
     punch
    -0.26
    SUCCESS
    -0.25
    æĮĩ
    -0.25
     Belt
    -0.25
    ,âĢ¦↵↵
    -0.25
     hasattr
    -0.24
     success
    -0.24
    çļĦäºĭå®ŀ
    -0.24
    POSITIVE LOGITS
    åħ¨åľº
    0.27
     initialValue
    0.26
    iban
    0.25
    åĿİ
    0.25
    è¯łéĩĬ
    0.25
    ene
    0.25
    ogonal
    0.25
    aff
    0.24
    Pipe
    0.24
    0.24
    Activations Density 1.328%

    No Known Activations