© 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. 20-TRANSCODER-HP
    7. 130096
    Prev
    Next
    INDEX
    Explanations

    Writing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Exhaust
    -0.26
    æķħ
    -0.25
    åıij表äºĨ
    -0.25
    è®®
    -0.24
    æµİ
    -0.24
    stadt
    -0.24
    pisa
    -0.23
    éĥ½æĥ³
    -0.23
    éĢĨ
    -0.23
     are
    -0.23
    POSITIVE LOGITS
    sing
    0.27
    "log
    0.27
     logs
    0.27
    raquo
    0.27
     harmless
    0.26
     blot
    0.26
     paramInt
    0.25
    AYS
    0.25
    ody
    0.24
     cent
    0.24
    Activations Density 7.269%

    No Known Activations