© 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. 30-TRANSCODER-HP
    7. 96603
    Prev
    Next
    INDEX
    Explanations

    PHP code/errors

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     str
    -0.36
     $_
    -0.35
    =str
    -0.31
    strtolower
    -0.30
     PDO
    -0.30
    isset
    -0.30
     sk
    -0.30
    åħī
    -0.30
     pow
    -0.28
    mysqli
    -0.28
    POSITIVE LOGITS
    Bundle
    0.32
     Bundle
    0.29
     Proto
    0.26
    bundle
    0.26
     Circus
    0.25
    lectron
    0.23
    ffield
    0.23
    amide
    0.23
    åīįç½®
    0.23
    .va
    0.23
    Activations Density 1.217%

    No Known Activations