© 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. 88143
    Prev
    Next
    INDEX
    Explanations

    forum/blog post excerpts

    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
     Brexit
    -0.30
    ABCDEFGHI
    -0.28
    åĽºå®ļèµĦ产
    -0.27
    ä¸įçŁ¥
    -0.26
    Instead
    -0.25
    onian
    -0.25
    (";
    -0.24
    stre
    -0.24
    (Self
    -0.24
     Instead
    -0.24
    POSITIVE LOGITS
    esome
    0.26
    èŀºæĹĭ
    0.26
    ä»Ģä¹Īåij¢
    0.26
     Charg
    0.25
    æ¿Ĥ
    0.25
    ier
    0.24
    èĪ³
    0.24
    à¹Ģศ
    0.24
    Bas
    0.24
     basal
    0.23
    Activations Density 21.388%

    No Known Activations