© 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. 107880
    Prev
    Next
    INDEX
    Explanations

    Formal/technical 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
    ç«Ļ
    -0.28
      ↵
    -0.25
     remot
    -0.25
     gib
    -0.24
     veget
    -0.24
    è¯ľ
    -0.24
    ACP
    -0.24
    åºļ
    -0.24
    }];↵
    -0.23
      ↵↵
    -0.23
    POSITIVE LOGITS
    ottom
    0.30
     unseen
    0.30
     Agree
    0.25
    uhn
    0.25
    ramid
    0.24
    Hier
    0.24
    collapse
    0.24
    cept
    0.24
     unfolded
    0.24
     taxed
    0.24
    Activations Density 14.146%

    No Known Activations