© 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. 96666
    Prev
    Next
    INDEX
    Explanations

    Code and machine learning

    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
     dut
    -0.29
    uum
    -0.29
     seasons
    -0.25
     dich
    -0.25
     dossier
    -0.24
    PROP
    -0.24
    jour
    -0.24
    движ
    -0.24
    hta
    -0.24
    åľ¨æİ¥åıĹ
    -0.23
    POSITIVE LOGITS
    å¾Ĺå¾Ī
    0.28
    å¼ĢæĶ¾
    0.28
    äºĨ
    0.26
    çIJĨæĢ§
    0.26
    predicted
    0.26
    ç³ľ
    0.26
    é¢Ħæµĭ
    0.25
    antic
    0.25
    é¢ĦæľŁ
    0.25
    ULD
    0.25
    Activations Density 0.002%

    No Known Activations