© 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. 33-TRANSCODER-HP
    7. 53073
    Prev
    Next
    INDEX
    Explanations

    contest, loss, or channels

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    èĬ±
    -0.27
    åĽłåľ°
    -0.25
    eling
    -0.25
    favorites
    -0.23
    .Batch
    -0.23
     Institutes
    -0.23
    esis
    -0.23
     thá»ijng
    -0.23
    oice
    -0.23
     Claus
    -0.23
    POSITIVE LOGITS
    ilver
    0.30
    æįŁåĿı
    0.27
    jen
    0.25
    lag
    0.25
    åĿİ
    0.25
    åıĮè¾¹
    0.24
     prost
    0.24
    æĿĥéĻIJ
    0.24
    ghan
    0.24
    Attempt
    0.24
    Activations Density 0.057%

    No Known Activations