© 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. 12-TRANSCODER-HP
    7. 23743
    Prev
    Next
    INDEX
    Explanations

    Code/technical content

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æµļ
    -0.28
    éĴĿ
    -0.28
    ("/",
    -0.28
     ì¹
    -0.26
    sun
    -0.26
    æĴĩ
    -0.26
    èIJ±
    -0.25
    ax
    -0.25
    cp
    -0.25
    mid
    -0.25
    POSITIVE LOGITS
    çī©çIJĨåѦ
    0.46
     physics
    0.44
    physics
    0.41
     physicists
    0.40
     Physics
    0.38
     physicist
    0.38
    .physics
    0.34
     photon
    0.34
     electromagnetic
    0.33
    åĬĽåѦ
    0.32
    Activations Density 0.000%

    No Known Activations