© 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. 89018
    Prev
    Next
    INDEX
    Explanations

    Abbreviations/Research

    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
    éĤ£éº¼
    -0.28
    åIJĥäºı
    -0.25
    *S
    -0.25
    inn
    -0.25
    请èģĶç³»
    -0.25
    irl
    -0.24
    ä¸Ĭ游
    -0.24
    æĺ¯ä¸Ģéĥ¨
    -0.24
     sticky
    -0.23
    rou
    -0.23
    POSITIVE LOGITS
    éļĪ
    0.27
    igo
    0.24
    yaw
    0.24
    lke
    0.23
     Yard
    0.23
    igor
    0.23
    -yard
    0.23
    obar
    0.23
    -demand
    0.23
    halt
    0.23
    Activations Density 76.939%

    No Known Activations