© 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. 25-TRANSCODER-HP
    7. 108723
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ointment
    -0.27
    ricks
    -0.27
    ece
    -0.25
     salute
    -0.24
     feast
    -0.24
    éħ·
    -0.24
    ".↵↵
    -0.24
    ogi
    -0.24
     reel
    -0.23
    临
    -0.23
    POSITIVE LOGITS
      ↵
    0.36
    :↵
    0.29
     ↵
    0.28
    ;↵
    0.28
    ï¼Ľ↵
    0.27
    åĿİ
    0.27
    ï¼ļ↵
    0.26
    erna
    0.26
    Âł↵
    0.26
       ↵
    0.25
    Activations Density 1.605%

    No Known Activations