© 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. 88523
    Prev
    Next
    INDEX
    Explanations

    formal documentation

    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
    çķĮçļĦ
    -0.28
    åı¯èĥ½åŃĺåľ¨
    -0.25
     haus
    -0.24
    缮åīįåĽ½åĨħ
    -0.24
    é¢ģåıij
    -0.24
    çķĮ
    -0.23
     strategies
    -0.23
     phạm
    -0.23
    uation
    -0.23
    éĩıåĮĸ
    -0.23
    POSITIVE LOGITS
    lesc
    0.26
     Analyst
    0.25
    ä¸Ģæ³¢
    0.25
    ники
    0.24
    çĭ¬ç«ĭ
    0.24
     Independence
    0.24
     Jin
    0.23
     Mesh
    0.23
     Bradford
    0.23
    деж
    0.23
    Activations Density 1.172%

    No Known Activations