© 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. 88714
    Prev
    Next
    INDEX
    Explanations

    Code identifiers

    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
    ÂŃi
    -0.28
    ihan
    -0.28
    ç½ijå°ıç¼ĸ
    -0.26
    éķ¿çŁŃ
    -0.26
    天ä¸ĭ
    -0.26
    Bound
    -0.25
    æľªçŁ¥
    -0.24
    æİ¥åıĹ
    -0.24
    ä¸įåľ¨
    -0.24
    æľĢåħĪ
    -0.24
    POSITIVE LOGITS
    çĽ
    0.30
    duce
    0.27
    hic
    0.26
    h
    0.25
    ewood
    0.24
    arily
    0.24
     Chancellor
    0.24
    aba
    0.24
     cylinders
    0.24
    hee
    0.24
    Activations Density 33.796%

    No Known Activations