© 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. 88140
    Prev
    Next
    INDEX
    Explanations

    Numbers and parentheses

    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
    代åĬŀ
    -0.30
    steen
    -0.29
    nte
    -0.27
    ard
    -0.26
     Trident
    -0.26
    姣
    -0.26
    ertz
    -0.26
    åĹħ
    -0.25
    ensed
    -0.24
    nton
    -0.24
    POSITIVE LOGITS
    ulos
    0.27
    haul
    0.27
    ä½ķå¿ħ
    0.26
    iane
    0.25
    ifikasi
    0.25
    uild
    0.25
    ç»ĵ
    0.25
     ?");↵
    0.24
    ****↵
    0.24
    -inf
    0.24
    Activations Density 3.776%

    No Known Activations