© 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. 88246
    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
    Negative Logits
    ishment
    -0.29
    /lists
    -0.28
    å¦Ĥæŀľä¸įæĺ¯
    -0.25
    icut
    -0.24
    åĪĨéĴŁåIJİ
    -0.24
    å³Ļ
    -0.23
    tlement
    -0.23
    minating
    -0.23
     lovers
    -0.23
    è¾Ľè¾£
    -0.23
    POSITIVE LOGITS
    storm
    0.26
    åı©
    0.25
    æĬ½
    0.25
    称为
    0.25
    deal
    0.25
    cpy
    0.24
    pler
    0.24
    (Activity
    0.24
    æıĨ
    0.24
    ome
    0.23
    Activations Density 7.349%

    No Known Activations

    This feature has no known activations.