© 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. 89992
    Prev
    Next
    INDEX
    Explanations

    Code/math/technical

    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
    Not in Any Lists

    No Comments

    Negative Logits
    luent
    -0.26
    listing
    -0.25
    inar
    -0.25
    union
    -0.25
    ioxide
    -0.25
    поÑĤ
    -0.24
    èµĦæł¼
    -0.24
    primary
    -0.24
    åįģ个
    -0.23
    role
    -0.23
    POSITIVE LOGITS
     flor
    0.30
     gent
    0.26
    unken
    0.25
     Gent
    0.25
     tac
    0.25
    è¶³å¤ŁçļĦ
    0.24
    emento
    0.24
     apro
    0.24
    -addons
    0.24
     SY
    0.23
    Activations Density 6.623%

    No Known Activations