© 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. 96049
    Prev
    Next
    INDEX
    Explanations

    technical document excerpts

    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
     pros
    -0.25
    ]['
    -0.25
     ster
    -0.25
     announces
    -0.24
    abee
    -0.24
    å°Ŀ
    -0.24
    Unc
    -0.24
     unc
    -0.24
     Pro
    -0.24
    ä¼°åĢ¼
    -0.24
    POSITIVE LOGITS
    åŁ¤
    0.27
    .clip
    0.25
    DOT
    0.25
    客车
    0.24
    todo
    0.24
    客人
    0.24
     yardımcı
    0.24
    ForeignKey
    0.24
    baar
    0.24
     guerr
    0.23
    Activations Density 0.785%

    No Known Activations