© 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. 31-TRANSCODER-HP
    7. 19403
    Prev
    Next
    INDEX
    Explanations

    say "rotein" words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æ´ĭ
    -0.26
     Attr
    -0.25
    asha
    -0.25
    å¾Ń
    -0.25
    èijŃ
    -0.25
    çħĮ
    -0.25
     attr
    -0.25
    Ny
    -0.24
    åijĭ
    -0.24
    Attr
    -0.23
    POSITIVE LOGITS
    èĽĭçĻ½
    0.98
     protein
    0.96
    èĽĭçĻ½è´¨
    0.93
     Prote
    0.92
    Pro
    0.89
     proteins
    0.88
     prote
    0.88
    protein
    0.84
    -pro
    0.83
     prot
    0.82
    Activations Density 11.598%

    No Known Activations