© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 88773
    Prev
    Next
    INDEX
    Explanations

    code/programming language

    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
     held
    -0.29
    oks
    -0.26
     skipped
    -0.25
    ä¾Ŀæ³ķ追究
    -0.25
    bbing
    -0.25
     toss
    -0.25
     licensors
    -0.25
     Harmony
    -0.24
    _instructions
    -0.24
    ffb
    -0.24
    POSITIVE LOGITS
    opoly
    0.31
    erman
    0.29
    ittel
    0.26
    åĤ
    0.25
    _asc
    0.25
    utzer
    0.25
     Resolver
    0.25
    ermann
    0.24
    autoload
    0.24
    elves
    0.24
    Activations Density 45.665%

    No Known Activations