© 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. 89024
    Prev
    Next
    INDEX
    Explanations

    programming and code

    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
    .nano
    -0.26
    rium
    -0.26
     barr
    -0.25
    让çĶ¨æĪ·
    -0.25
    ÑģпоÑĢ
    -0.24
    Ắ
    -0.24
    earch
    -0.24
    resources
    -0.24
    èĩªåĪ¶
    -0.24
     conserv
    -0.24
    POSITIVE LOGITS
    ä¸Ģåij³
    0.28
    ANA
    0.26
    尽管
    0.25
     Andrea
    0.25
    itur
    0.24
     Zus
    0.24
    ä¸Ģ头
    0.24
    cox
    0.24
    IDO
    0.23
    ç§ĺ书éķ¿
    0.23
    Activations Density 67.428%

    No Known Activations